🔭Data Science: The Truth in Models (Just the Quotes)

"A model, like a novel, may resonate with nature, but it is not a ‘real’ thing. Like a novel, a model may be convincing - it may ‘ring true’ if it is consistent with our experience of the natural world. But just as we may wonder how much the characters in a novel are drawn from real life and how much is artifice, we might ask the same of a model: How much is based on observation and measurement of accessible phenomena, how much is convenience? Fundamentally, the reason for modeling is a lack of full access, either in time or space, to the phenomena of interest." (Kenneth Belitz, Science, Vol. 263, 1944)

"Exact truth of a null hypothesis is very unlikely except in a genuine uniformity trial." (David R Cox, "Some problems connected with statistical inference", Annals of Mathematical Statistics 29, 1958)

"[…] no models are [true] = not even the Newtonian laws. When you construct a model you leave out all the details which you, with the knowledge at your disposal, consider inessential. […] Models should not be true, but it is important that they are applicable, and whether they are applicable for any given purpose must of course be investigated. This also means that a model is never accepted finally, only on trial." (Georg Rasch, "Probabilistic Models for Some Intelligence and Attainment Tests", 1960)

"The validation of a model is not that it is 'true' but that it generates good testable hypotheses relevant to important problems." (Richard Levins, "The Strategy of Model Building in Population Biology", 1966)

"A theory has only the alternative of being right or wrong. A model has a third possibility: it may be right, but irrelevant." (Manfred Eigen, 1973)

"Models, of course, are never true, but fortunately it is only necessary that they be useful. For this it is usually needful only that they not be grossly wrong. I think rather simple modifications of our present models will prove adequate to take account of most realities of the outside world. The difficulties of computation which would have been a barrier in the past need not deter us now." (George E P Box, "Some Problems of Statistics and Everyday Life", Journal of the American Statistical Association, Vol. 74 (365), 1979)

"The purpose of an experiment is to answer questions. The truth of this seems so obvious, that it would not be worth emphasizing were it not for the fact that the results of many experiments are interpreted and presented with little or no reference to the questions that were asked in the first place."  (Thomas M Little, "Interpretation and presentation of results", Hortscience 16, 1981)

"The fact that [the model] is an approximation does not necessarily detract from its usefulness because models are approximations. All models are wrong, but some are useful." (George Box, 1987)

"A null hypothesis that yields under two different treatments have identical expectations is scarcely very plausible, and its rejection by a significance test is more dependent upon the size of an experiment than upon its untruth." (David J Finney, "Was this in your statistics textbook?", Experimental Agriculture 24, 1988)

"Statistical models for data are never true. The question whether a model is true is irrelevant. A more appropriate question is whether we obtain the correct scientific conclusion if we pretend that the process under study behaves according to a particular statistical model." (Scott Zeger, "Statistical reasoning in epidemiology", American Journal of Epidemiology, 1991)

"The motivation for any action on outliers must be to improve interpretation of data without ignoring unwelcome truth. To remove bad and untrustworthy data is a laudable ambition, but naive and untested rules may bring harm rather than benefit." (David Finney, "Calibration Guidelines Challenge Outlier Practices", The American Statistician Vol 60 (4), 2006) 

"You might say that there’s no reason to bother with model checking since all models are false anyway. I do believe that all models are false, but for me the purpose of model checking is not to accept or reject a model, but to reveal aspects of the data that are not captured by the fitted model." (Andrew Gelman, "Some thoughts on the sociology of statistics", 2007)

"If students have students have no experience with hands-on [telescope] observing, they may take all data as ‘truth’ without having an understanding of how the data are obtained and what could potentially go wrong in that process, so I think it becomes crucially important to give a glimpse of what’s happening behind the scenes at telescopes, so they can be appropriately skeptical users of data in the future." (Colette Salyk, Sky & Telescope, 2022)

"On a final note, we would like to stress the importance of design, which often does not receive the attention it deserves. Sometimes, the large number of modeling options for spatial analysis may raise the false impression that design does not matter, and that a sophisticated analysis takes care of everything. Nothing could be further from the truth." (Hans-Peter Piepho et al, "Two-dimensional P-spline smoothing for spatial analysis of plant breeding trials", “Biometrical Journal”, 2022)

🔭Data Science: Outliers (Just the Quotes)

"An observation with an abnormally large residual will be referred to as an outlier. Other terms in English are 'wild', 'straggler', 'sport' and 'maverick'; one may also speak of a 'discordant', 'anomalous' or 'aberrant' observation." (Francis J Anscombe, "Rejection of Outliers", Technometrics Vol. 2, 1960)

"One sufficiently erroneous reading can wreck the whole of a statistical analysis, however many observations there are." (Francis J Anscombe, "Rejection of Outliers", Technometrics Vol. 2, 1960)

"The fact that something is far-fetched is no reason why it should not be true; it cannot be as far-fetched as the fact that something exists." (Celia Green, "The Decline and Fall of Science", 1976)

"When the statistician looks at the outside world, he cannot, for example, rely on finding errors that are independently and identically distributed in approximately normal distributions. In particular, most economic and business data are collected serially and can be expected, therefore, to be heavily serially dependent. So is much of the data collected from the automatic instruments which are becoming so common in laboratories these days. Analysis of such data, using procedures such as standard regression analysis which assume independence, can lead to gross error. Furthermore, the possibility of contamination of the error distribution by outliers is always present and has recently received much attention. More generally, real data sets, especially if they are long, usually show inhomogeneity in the mean, the variance, or both, and it is not always possible to randomize." (George E P Box, "Some Problems of Statistics and Everyday Life", Journal of the American Statistical Association, Vol. 74 (365), 1979)

"A good description of the data summarizes the systematic variation and leaves residuals that look structureless. That is, the residuals exhibit no patterns and have no exceptionally large values, or outliers. Any structure present in the residuals indicates an inadequate fit. Looking at the residuals laid out in an overlay helps to spot patterns and outliers and to associate them with their source in the data." (Christopher H Schrnid, "Value Splitting: Taking the Data Apart", 1991)

"So we pour in data from the past to fuel the decision-making mechanisms created by our models, be they linear or nonlinear. But therein lies the logician's trap: past data from real life constitute a sequence of events rather than a set of independent observations, which is what the laws of probability demand. [...] It is in those outliers and imperfections that the wildness lurks." (Peter L Bernstein, "Against the Gods: The Remarkable Story of Risk", 1996)

"I have often thought that outliers contain more information than the model." (Arnold Goodman,  [Joint Statistical Meetings] 2005)

"I have often thought that outliers contain more information than the model." (Arnold Goodman,  [Joint Statistical Meetings] 2005)"The finding of an outlier is not necessarily a discovery of a bad or misleading datum that may contaminate the data, but it may amount to a comment on the validity of distributional assumptions inherent in the form of analysis that is contemplated." (David Finney, "Calibration Guidelines Challenge Outlier Practices", The American Statistician Vol 60 (4), 2006)

"One cautious approach is represented by Bernoulli’s more conservative outlook. If there are very strong reasons for believing that an observation has suffered an accident that made the value in the data-file thoroughly untrustworthy, then reject it; in the absence of clear evidence that an observation, identified by formal rule as an outlier, is unacceptable then retain it unless there is lack of trust that the laboratory obtaining it is conscientiously operated by able persons who have [... ] taken every care.'" (David Finney, "Calibration Guidelines Challenge Outlier Practices", The American Statistician Vol 60 (4), 2006)

"Why is a particular record or measurement classed as an outlier? Among all who handle and interpret statistical data, the word has long been in common use as an epithet for any item among a dataset of N that departs markedly from the broad pattern of the set." (David Finney, "Calibration Guidelines Challenge Outlier Practices", The American Statistician Vol 60 (4), 2006)

"All this discussion of deleting the outliers is completely backwards. In my work, I usually throw away all the good data, and just analyze the outliers." (Anon, The American Statistician Vol 61(3), 2007)

"Before discarding a data point one should investigate the possible reasons for this faulty data value." (Manfred Drosg, "Dealing with Uncertainties: A Guide to Error Analysis", 2007)

"If there is an outlier there are two possibilities: The model is wrong – after all, a theory is the basis on which we decide whether a data point is an outlier (an unexpected value) or not. The value of the data point is wrong because of a failure of the apparatus or a human mistake. There is a third possibility, though: The data point might not be an actual  outlier, but part of a (legitimate) statistical fluctuation." (Manfred Drosg, "Dealing with Uncertainties: A Guide to Error Analysis", 2007)

"Outliers or flyers are those data points in a set that do not quite fit within the rest of the data, that agree with the model in use. The uncertainty of such an outlier is seemingly too small. The discrepancy between outliers and the model should be subject to thorough examination and should be given much thought. Isolated data points, i.e., data points that are at some distance from the bulk of the data are not outliers if their values are in agreement with the model in use." (Manfred Drosg, "Dealing with Uncertainties: A Guide to Error Analysis", 2007)

"The inability to predict outliers implies the inability to predict the course of history." (Nassim N Taleb, "The Black Swan", 2007)

"Given the important role that correlation plays in structural equation modeling, we need to understand the factors that affect establishing relationships among multivariable data points. The key factors are the level of measurement, restriction of range in data values (variability, skewness, kurtosis), missing data, nonlinearity, outliers, correction for attenuation, and issues related to sampling variation, confidence intervals, effect size, significance, sample size, and power." (Randall E Schumacker & Richard G Lomax, "A Beginner’s Guide to Structural Equation Modeling" 3rd Ed., 2010)

"Need to consider outliers as they can affect statistics such as means, standard deviations, and correlations. They can either be explained, deleted, or accommodated (using either robust statistics or obtaining additional data to fill-in). Can be detected by methods such as box plots, scatterplots, histograms or frequency distributions." (Randall E Schumacker & Richard G Lomax, "A Beginner’s Guide to Structural Equation Modeling" 3rd Ed., 2010)

"Outliers or influential data points can be defined as data values that are extreme or atypical on either the independent (X variables) or dependent (Y variables) variables or both. Outliers can occur as a result of observation errors, data entry errors, instrument errors based on layout or instructions, or actual extreme values from self-report data. Because outliers affect the mean, the standard deviation, and correlation coefficient values, they must be explained, deleted, or accommodated by using robust statistics." (Randall E Schumacker & Richard G Lomax, "A Beginner’s Guide to Structural Equation Modeling" 3rd Ed., 2010)

"There are several key issues in the field of statistics that impact our analyses once data have been imported into a software program. These data issues are commonly referred to as the measurement scale of variables, restriction in the range of data, missing data values, outliers, linearity, and nonnormality." (Randall E Schumacker & Richard G Lomax, "A Beginner’s Guide to Structural Equation Modeling" 3rd Ed., 2010)

"After you visualize your data, there are certain things to look for […]: increasing, decreasing, outliers, or some mix, and of course, be sure you’re not mixing up noise for patterns. Also note how much of a change there is and how prominent the patterns are. How does the difference compare to the randomness in the data? Observations can stand out because of human or mechanical error, because of the uncertainty of estimated values, or because there was a person or thing that stood out from the rest. You should know which it is." (Nathan Yau, "Data Points: Visualization That Means Something", 2013)

"A major advantage of probabilistic models is that they can be easily applied to virtually any data type (or mixed data type), as long as an appropriate generative model is available for each mixture component. [...] A downside of probabilistic models is that they try to fit the data to a particular kind of distribution, which may often not be appropriate for the underlying data. Furthermore, as the number of model parameters increases, over-fitting becomes more common. In such cases, the outliers may fit the underlying model of normal data. Many parametric models are also harder to interpret in terms of intensional knowledge, especially when the parameters of the model cannot be intuitively presented to an analyst in terms of underlying attributes. This can defeat one of the important purposes of anomaly detection, which is to provide diagnostic understanding of the abnormal data generative process." (Charu C Aggarwal, "Outlier Analysis", 2013)

"An attempt to use the wrong model for a given data set is likely to provide poor results. Therefore, the core principle of discovering outliers is based on assumptions about the structure of the normal patterns in a given data set. Clearly, the choice of the 'normal' model depends highly upon the analyst’s understanding of the natural data patterns in that particular domain." (Charu C Aggarwal, "Outlier Analysis", 2013)

"Typically, most outlier detection algorithms use some quantified measure of the outlierness of a data point, such as the sparsity of the underlying region, nearest neighbor based distance, or the fit to the underlying data distribution. Every data point lies on a continuous spectrum from normal data to noise, and finally to anomalies [...] The separation of the different regions of this spectrum is often not precisely defined, and is chosen on an ad-hoc basis according to application-specific criteria. Furthermore, the separation between noise and anomalies is not pure, and many data points created by a noisy generative process may be deviant enough to be interpreted as anomalies on the basis of the outlier score. Thus, anomalies will typically have a much higher outlier score than noise, but this is not a distinguishing factor between the two as a matter of definition. Rather, it is the interest of the analyst, which regulates the distinction between noise and an anomaly." (Charu C Aggarwal, "Outlier Analysis", 2013) 

"What is good visualization? It is a representation of data that helps you see what you otherwise would have been blind to if you looked only at the naked source. It enables you to see trends, patterns, and outliers that tell you about yourself and what surrounds you. The best visualization evokes that moment of bliss when seeing something for the first time, knowing that what you see has been right in front of you, just slightly hidden. Sometimes it is a simple bar graph, and other times the visualization is complex because the data requires it." (Nathan Yau, "Data Points: Visualization That Means Something", 2013)

"When data is not normal, the reason the formulas are working is usually the central limit theorem. For large sample sizes, the formulas are producing parameter estimates that are approximately normal even when the data is not itself normal. The central limit theorem does make some assumptions and one is that the mean and variance of the population exist. Outliers in the data are evidence that these assumptions may not be true. Persistent outliers in the data, ones that are not errors and cannot be otherwise explained, suggest that the usual procedures based on the central limit theorem are not applicable." (DeWayne R Derryberry, "Basic data analysis for time series with R", 2014)

"When we find data quality issues due to valid data during data exploration, we should note these issues in a data quality plan for potential handling later in the project. The most common issues in this regard are missing values and outliers, which are both examples of noise in the data." (John D Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, worked examples, and case studies", 2015)

"Whatever actually happened, outliers need to be investigated not omitted. Try to understand what caused some observations to be different from the bulk of the observations. If you understand the reasons, you are then in a better position to judge whether the points can legitimately removed from the data set, or whether you’ve just discovered something new and interesting. Never remove a point just because it is weird." (Rob J Hyndman, "Omitting outliers", 2016)

"There are a lot of statistical methods looking at whether an outlier should be deleted[...] I don’t endorse any of them." (Barry Nussbaum, "Significance", 2017)

"Outliers make it very hard to give an intuitive interpretation of the mean, but in fact, the situation is even worse than that. For a real‐world distribution, there always is a mean (strictly speaking, you can define distributions with no mean, but they’re not realistic), and when we take the average of our data points, we are trying to estimate that mean. But when there are massive outliers, just a single data point is likely to dominate the value of the mean and standard deviation, so much more data is required to even estimate the mean, let alone make sense of it." (Field Cady, "The Data Science Handbook", 2017)

"[...] data often has some errors, outliers and other strange values, but these do not necessarily need to be individually identified and excluded. It also points to the benefits of using summary measures that are not unduly affected by odd observations [...] are known as robust measures, and include the median and the inter-quartile range." (David Spiegelhalter, "The Art of Statistics: Learning from Data", 2019)

"When visuals are applied to data, they can enlighten the audience to insights that they wouldn’t see without charts or graphs. Many interesting patterns and outliers in the data would remain hidden in the rows and columns of data tables without the help of data visualizations. They connect with our visual nature as human beings and impart knowledge that couldn’t be obtained as easily using other approaches that involve just words or numbers." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"An outlier is a data point that is far away from other observations in your data. It may be due to random variability in the data, measurement error, or an actual anomaly. Outliers are both an opportunity and a warning. They potentially give you something very interesting to talk about, or they may signal that something is wrong in the data." (Jonathan Schwabish, "Better Data Visualizations: A guide for scholars, researchers, and wonks", 2021)

"Visualizations can remove the background noise from enormous sets of data so that only the most important points stand out to the intended audience. This is particularly important in the era of big data. The more data there is, the more chance for noise and outliers to interfere with the core concepts of the data set." (Kate Strachnyi, "ColorWise: A Data Storyteller’s Guide to the Intentional Use of Color", 2023)

"I don’t see the logic of rejecting data just because they seem incredible." (Fred Hoyle)

"In almost every true series of observations, some are found, which differ so much from the others as to indicate some abnormal source of error not contemplated in the theoretical discussions, and the introduction of which into the investigations can only serve, in the present state of science, to perplex and mislead the inquirer." (Benjamin Peirce, The Astronomical Journal)

"Treat outliers like children. Correct them when necessary, but never throw them out." (Anon)

🔭Data Science: Data Stories (Just the Quotes)

"Stories have a marvelous way of focusing our attention and helping us to discern why the data presented is important or relevant to some part of our lives. It is only inside of a context that data is meaningful, and using the data as part of a story is an excellent way of allowing the data to make a lasting impact. The most effective information visualizations will make themselves a pivotal point in a story or narrative within the viewers’ (or users’) minds." (Matthias Shapiro, "Once Upon a Stacked Time Series", [in "Beautiful Visualization"] 2010)

"A useful way to look at a data visualization challenge is to recognize that we are actually seeking to reduce choices. This is achieved through recognizing influential factors, by considering the desired function and tone of our work, familiarizing with our data and identifying stories. We are building clarity through selection and rejection. We are reducing the problem by enhancing our clarity." (Andy Kirk, "Data Visualization: A successful design process", 2012)

"At its best, a static visualization is like a powerful photograph - a carefully conceived, arranged, and executed vision that manages to portray the sequence or motion of a story without the actual deployment of movement." (Andy Kirk, "Data Visualization: A successful design process", 2012)

"Data stories are a subset of the much broader concept (or buzzword) of storytelling. […] Stories, or narratives, are useful in data visualization because they force us to recognize the limited value of a single chart in a complex environment. Stories also force us to recognize the need for a better integration of our displays, as we move away from strings of siloed charts." (Jorge Camões, "Data at Work: Best practices for creating effective charts and information graphics in Microsoft Excel", 2016)

"A data story starts out like any other story, with a beginning and a middle. However, the end should never be a fixed event, but rather a set of options or questions to trigger an action from the audience. Never forget that the goal of data storytelling is to encourage and energize critical thinking for business decisions." (James Richardson, 2017)

"All human storytellers bring their subjectivity to their narratives. All have bias, and possibly error. Acknowledging and defusing that bias is a vital part of successfully using data stories. By debating a data story collaboratively and subjecting it to critical thinking, organizations can get much higher levels of engagement with data and analytics and impact their decision making much more than with reports and dashboards alone." (James Richardson, 2017)

"A random collection of interesting but disconnected facts will lack the unifying theme to become a data story - it may be informative, but it won’t be insightful." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Analysis is a two-step process that has an exploratory and an explanatory phase. In order to create a powerful data story, you must effectively transition from data discovery (when you’re finding insights) to data communication (when you’re explaining them to an audience). If you don’t properly traverse these two phases, you may end up with something that resembles a data story but doesn’t have the same effect. Yes, it may have numbers, charts, and annotations, but because it’s poorly formed, it won’t achieve the same results." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Before you can even consider creating a data story, you must have a meaningful insight to share. One of the essential attributes of a data story is a central or main insight. Without a main point, your data story will lack purpose, direction, and cohesion. A central insight is the unifying theme (telos appeal) that ties your various findings together and guides your audience to a focal point or climax for your data story. However, when you have an increasing amount of data at your disposal, insights can be elusive. The noise from irrelevant and peripheral data can interfere with your ability to pinpoint the important signals hidden within its core." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Data storytelling gives your insight the best opportunity to capture attention, be understood, be remembered, and be acted on. An effective data story helps your insight reach its full potential: inspiring others to act and drive change." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Data storytelling involves the skillful combination of three key elements: data, narrative, and visuals. Data is the primary building block of every data story. It may sound simple, but a data story should always find its origin in data, and data should serve as the foundation for the narrative and visual elements of your story." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Even with a solid narrative and insightful visuals, a data story cannot overcome a weak data foundation. As the master architect, builder, and designer of your data story, you play an instrumental role in ensuring its truthfulness, quality, and effectiveness. Because you are responsible for pouring the data foundation and framing the narrative structure of your data story, you need to be careful during the analysis process. Because all of the data is being processed and interpreted by you before it is shared with others, it can be exposed to cognitive biases and logical fallacies that distort or weaken the data foundation of your story." (Brent Dykes, "Effective Data Storytelling: How to Drive Change with Data, Narrative and Visuals", 2019)

"Numbers are ideal vehicles for promulgating bullshit. They feel objective, but are easily manipulated to tell whatever story one desires. Words are clearly constructs of human minds, but numbers? Numbers seem to come directly from Nature herself. We know words are subjective. We know they are used to bend and blur the truth. Words suggest intuition, feeling, and expressivity. But not numbers. Numbers suggest precision and imply a scientific approach. Numbers appear to have an existence separate from the humans reporting them." (Carl T Bergstrom & Jevin D West, "Calling Bullshit: The Art of Skepticism in a Data-Driven World", 2020)

"So what does it mean to tell an honest story? Numbers should be presented in ways that allow meaningful comparisons." (Carl T Bergstrom & Jevin D West, "Calling Bullshit: The Art of Skepticism in a Data-Driven World", 2020)

"To tell an honest story, it is not enough for numbers to be correct. They need to be placed in an appropriate context so that a reader or listener can properly interpret them." (Carl T Bergstrom & Jevin D West, "Calling Bullshit: The Art of Skepticism in a Data-Driven World", 2020)

"Good data stories have three key components: data, narrative, and visuals. [...] The data part is fairly obvious - data has to be accurate for the correct insights to be achieved. The narrative has to give a voice to the data in simple language, turning each data point into a character in the story with its own tale to tell. The visuals are what we are most concerned about. They have to allow us to be able to find trends and patterns in our datasets and do so easily and specifically. The last thing we want is for the most important points to be buried in rows and columns." (Kate Strachnyi, "ColorWise: A Data Storyteller’s Guide to the Intentional Use of Color", 2023)

"One of the most challenging components of data visualization is that oftentimes the story you need to tell is complex, dynamic, and multidimensional. However, the standard tools we have are flat, static, and designed for paper. Decomposing the dynamic nature of the narrative embedded within your data into a storyboard format is one of the best ways to ensure your key points are effectively received by your intended audience." (Thomas Rhodes)

🔭Data Science: Scientific Method (Just the Quotes)

"There is no short cut to truth, no way to gain a knowledge of the universe except through the gateway of scientific method." (Karl Pearson, "The Grammar of Science", 1892)

“It must be gently but firmly pointed out that analogy is the very corner-stone of scientific method. A root-and-branch condemnation would invalidate any attempt to explain the unknown in terms of the known, and thus prune away every hypothesis.” (Archie E Heath, “On Analogy”, The Cambridge Magazine, 1918)

"The fundamental difference between engineering with and without statistics boils down to the difference between the use of a scientific method based upon the concept of laws of nature that do not allow for chance or uncertainty and a scientific method based upon the concepts of laws of probability as an attribute of nature." (Walter A Shewhart, 1940)

"When the number of factors coming into play in a phenomenological complex is too large, scientific method in most cases fails us. One need only think of the weather, in which case prediction even for a few days ahead is impossible. Nevertheless no one doubts that we are confronted with a causal connection whose causal components are in the main known to us. Occurrences in this domain are beyond the reach of exact prediction because of the variety of factors in operation, not because of any lack of order in nature." (Albert Einstein, "Science and Religion", 1941)

"The construction of hypotheses is a creative act of inspiration, intuition, invention; its essence is the vision of something new in familiar material. The process must be discussed in psychological, not logical, categories; studied in autobiographies and biographies, not treatises on scientific method; and promoted by maxim and example, not syllogism or theorem." (Milton Friedman, "Essays in Positive Economics", 1953)

"Science cannot be based on dogma or authority of any kind, nor on any institution or revelation, unless indeed it be of the Book of Nature that lies open before our eyes. We need not dwell on the processes of acquiring knowledge by observation, experiment, and inductive and deductive reasoning. The study of scientific method both in theory and practice is of great importance. It is inherent in the philosophy that the record may be imperfect and the conceptions erroneous; the potential fallibility of our science is not only acknowledged but also insisted upon." (Sir Robert Robinson, "Science and the Scientist", Nature Vol. 176 (4479), 1955)

"Scientific method is the way to truth, but it affords, even in principle, no unique definition of truth. Any so-called pragmatic definition of truth is doomed to failure equally." (Willard v O Quine, "Word and Object", 1960)

"Model-making, the imaginative and logical steps which precede the experiment, may be judged the most valuable part of scientific method because skill and insight in these matters are rare. Without them we do not know what experiment to do. But it is the experiment which provides the raw material for scientific theory. Scientific theory cannot be built directly from the conclusions of conceptual models." (Herbert G Andrewartha, "Introduction to the Study of Animal Population", 1961)

"Statistics is the branch of scientific method which deals with the data obtained by counting or measuring the properties of populations of natural phenomena." (Sir Maurice G Kendall & Alan Stuart, "The Advanced Theory of Statistics", 1963)

"Observation, reason, and experiment make up what we call the scientific method. (Richard Feynman, "Mainly mechanics, radiation, and heat", 1963)

"The heart of the scientific method is the problem-hypothesis-test process. And, necessarily, the scientific method involves predictions. And predictions, to be useful in scientific methodology, must be subject to test empirically." (Paul Davies, "The Cosmic Blueprint: New Discoveries in Nature's Creative Ability to, Order the Universe", 1988)

"Scientists use mathematics to build mental universes. They write down mathematical descriptions - models - that capture essential fragments of how they think the world behaves. Then they analyse their consequences. This is called 'theory'. They test their theories against observations: this is called 'experiment'. Depending on the result, they may modify the mathematical model and repeat the cycle until theory and experiment agree. Not that it's really that simple; but that's the general gist of it, the essence of the scientific method." (Ian Stewart & Martin Golubitsky, "Fearful Symmetry: Is God a Geometer?", 1992)

"But our ways of learning about the world are strongly influenced by the social preconceptions and biased modes of thinking that each scientist must apply to any problem. The stereotype of a fully rational and objective ‘scientific method’, with individual scientists as logical (and interchangeable) robots, is self-serving mythology." (Stephen J Gould, "This View of Life: In the Mind of the Beholder", "Natural History", Vol. 103, No. 2, 1994)

"The traditional, scientific method for studying such systems is known as reductionism. Reductionism sees the parts as paramount and seeks to identify the parts, understand the parts and work up from an understanding of the parts to an understanding of the whole. The problem with this is that the whole often seems to take on a form that is not recognizable from the parts. The whole emerges from the interactions between the parts, which affect each other through complex networks of relationships. Once it has emerged, it is the whole that seems to give meaning to the parts and their interactions." (Michael C Jackson, "Systems Thinking: Creative Holism for Managers", 2003)

"Scientists pursue ideas in an ill-defined but effective way that is often called the scientific method. There is no strict rule of procedure that will lead you from a good idea to a Nobel prize or even to a publishable discovery. Some scientists are meticulously careful; others are highly creative. The best scientists are probably both careful and creative. Although there are various scientific methods in use, a typical approach consists of a series of steps." (Peter Atkins et al, "Chemical Principles: The Quest for Insight" 6th ed., 2013)

"Science, at its core, is simply a method of practical logic that tests hypotheses against experience. Scientism, by contrast, is the worldview and value system that insists that the questions the scientific method can answer are the most important questions human beings can ask, and that the picture of the world yielded by science is a better approximation to reality than any other." (John M Greer, "After Progress: Reason and Religion at the End of the Industrial Age", 2015)

"The scientific method does not begin with the injunction to use reason and logic, and to obey the principle of sufficient reason and Occam’s razor. Instead, it begins with the word 'Observe'. In other words, if reality in itself is unobservable - which is of course the case - then the scientific method automatically fails to tell us a single thing about it." (Thomas Stark, "God Is Mathematics: The Proofs of the Eternal Existence of Mathematics", 2018)

"One of the severest tests of a scientific mind is to discern the limits of the legitimate application of the scientific method."  (James C Maxwell)

More quotes on "Scientific Method" at the-web-of-knowledge.blogspot.com. 

🔭Data Science: Noise (Just the Quotes)

"Information that is only partially structured (and therefore contains some 'noise' is fuzzy, inconsistent, and indistinct. Such imperfect information may be regarded as having merit only if it represents an intermediate step in structuring the information into a final meaningful form. If the partially Structured information remains in fuzzy form, it will create a state of dissatisfaction in the mind of the originator and certainly in the mind of the recipient. The natural desire is to continue structuring until clarity, simplicity, precision, and definitiveness are obtained." (Cecil H Meyers, "Handbook of Basic Graphs: A modern approach", 1970)

"To understand the need for structuring information, we should examine its opposite - nonstructured information. Nonstructured information may be thought of as exists and can be heard (or sensed with audio devices), but the mind attaches no rational meaning to the sound. In another sense, noise can be equated to writing a group of letters, numbers, and other symbols on a page without any design or key to their meaning. In such a situation, there is nothing the mind can grasp. Nonstructured information can be classified as useless, unless meaning exists somewhere in the jumble and a key can be found to unlock its hidden significance." (Cecil H Meyers, "Handbook of Basic Graphs: A modern approach", 1970)

"Neither noise nor information is predictable." (Ray Kurzweil, "The Age of Spiritual Machines: When Computers Exceed Human Intelligence", 1999)

"Data are collected as a basis for action. Yet before anyone can use data as a basis for action the data have to be interpreted. The proper interpretation of data will require that the data be presented in context, and that the analysis technique used will filter out the noise."  (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"Data are generally collected as a basis for action. However, unless potential signals are separated from probable noise, the actions taken may be totally inconsistent with the data. Thus, the proper use of data requires that you have simple and effective methods of analysis which will properly separate potential signals from probable noise." (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"No matter what the data, and no matter how the values are arranged and presented, you must always use some method of analysis to come up with an interpretation of the data. While every data set contains noise, some data sets may contain signals. Therefore, before you can detect a signal within any given data set, you must first filter out the noise." (Donald J Wheeler," Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"We analyze numbers in order to know when a change has occurred in our processes or systems. We want to know about such changes in a timely manner so that we can respond appropriately. While this sounds rather straightforward, there is a complication - the numbers can change even when our process does not. So, in our analysis of numbers, we need to have a way to distinguish those changes in the numbers that represent changes in our process from those that are essentially noise." (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"While all data contain noise, some data contain signals. Before you can detect a signal, you must filter out the noise." (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"The acquisition of information is a flow from noise to order - a process converting entropy to redundancy. During this process, the amount of information decreases but is compensated by constant re-coding. In the recoding the amount of information per unit increases by means of a new symbol which represents the total amount of the old. The maturing thus implies information condensation. Simultaneously, the redundance decreases, which render the information more difficult to interpret." (Lars Skyttner, "General Systems Theory: Ideas and Applications", 2001)

"In fact, an information theory that leaves out the issue of noise turns out to have no content." (Hans Christian von Baeyer, "Information, The New Language of Science", 2003)

"This phenomenon, common to chaos theory, is also known as sensitive dependence on initial conditions. Just a small change in the initial conditions can drastically change the long-term behavior of a system. Such a small amount of difference in a measurement might be considered experimental noise, background noise, or an inaccuracy of the equipment." (Greg Rae, Chaos Theory: A Brief Introduction, 2006)

"Data analysis is not generally thought of as being simple or easy, but it can be. The first step is to understand that the purpose of data analysis is to separate any signals that may be contained within the data from the noise in the data. Once you have filtered out the noise, anything left over will be your potential signals. The rest is just details." (Donald J Wheeler," Myths About Data Analysis", International Lean & Six Sigma Conference, 2012)

"Distinguishing the signal from the noise requires both scientific knowledge and self-knowledge." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Economists should study financial markets as they actually operate, not as they assume them to operate - observing the way in which information is actually processed, observing the serial correlations, bonanzas, and sudden stops, not assuming these away as noise around the edges of efficient and rational markets." (Adair Turner, "Economics after the Crisis: Objectives and means", 2012)

"Finding patterns is easy in any kind of data-rich environment; that's what mediocre gamblers do. The key is in determining whether the patterns represent signal or noise." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"The signal is the truth. The noise is what distracts us from the truth." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"Typically, most outlier detection algorithms use some quantified measure of the outlierness of a data point, such as the sparsity of the underlying region, nearest neighbor based distance, or the fit to the underlying data distribution. Every data point lies on a continuous spectrum from normal data to noise, and finally to anomalies [...] The separation of the different regions of this spectrum is often not precisely defined, and is chosen on an ad-hoc basis according to application-specific criteria. Furthermore, the separation between noise and anomalies is not pure, and many data points created by a noisy generative process may be deviant enough to be interpreted as anomalies on the basis of the outlier score. Thus, anomalies will typically have a much higher outlier score than noise, but this is not a distinguishing factor between the two as a matter of definition. Rather, it is the interest of the analyst, which regulates the distinction between noise and an anomaly." (Charu C Aggarwal, "Outlier Analysis", 2013)

"A complete data analysis will involve the following steps: (i) Finding a good model to fit the signal based on the data. (ii) Finding a good model to fit the noise, based on the residuals from the model. (iii) Adjusting variances, test statistics, confidence intervals, and predictions, based on the model for the noise." (DeWayne R Derryberry, "Basic data analysis for time series with R", 2014)

 "The random element in most data analysis is assumed to be white noise - normal errors independent of each other. In a time series, the errors are often linked so that independence cannot be assumed (the last examples). Modeling the nature of this dependence is the key to time series." (DeWayne R Derryberry, "Basic data analysis for time series with R", 2014)

"A signal is a useful message that resides in data. Data that isn’t useful is noise. […] When data is expressed visually, noise can exist not only as data that doesn’t inform but also as meaningless non-data elements of the display (e.g. irrelevant attributes, such as a third dimension of depth in bars, color variation that has no significance, and artificial light and shadow effects)." (Stephen Few, "Signal: Understanding What Matters in a World of Noise", 2015)

"Data contain descriptions. Some are true, some are not. Some are useful, most are not. Skillful use of data requires that we learn to pick out the pieces that are true and useful. [...] To find signals in data, we must learn to reduce the noise - not just the noise that resides in the data, but also the noise that resides in us. It is nearly impossible for noisy minds to perceive anything but noise in data." (Stephen Few, "Signal: Understanding What Matters in a World of Noise", 2015)

"When we find data quality issues due to valid data during data exploration, we should note these issues in a data quality plan for potential handling later in the project. The most common issues in this regard are missing values and outliers, which are both examples of noise in the data." (John D Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, worked examples, and case studies", 2015)

"Information theory leads to the quantification of the information content of the source, as denoted by entropy, the characterization of the information-bearing capacity of the communication channel, as related to its noise characteristics, and consequently the establishment of the relationship between the information content of the source and the capacity of the channel. In short, information theory provides a quantitative measure of the information contained in message signals and help determine the capacity of a communication system to transfer this information from source to sink over a noisy channel in a reliable fashion." (Ali Grami, "Information Theory", 2016)

"Repeated observations of the same phenomenon do not always produce the same results, due to random noise or error. Sampling errors result when our observations capture unrepresentative circumstances, like measuring rush hour traffic on weekends as well as during the work week. Measurement errors reflect the limits of precision inherent in any sensing device. The notion of signal to noise ratio captures the degree to which a series of observations reflects a quantity of interest as opposed to data variance. As data scientists, we care about changes in the signal instead of the noise, and such variance often makes this problem surprisingly difficult." (Steven S Skiena, "The Data Science Design Manual", 2017)

"Using noise (the uncorrelated variables) to fit noise (the residual left from a simple model on the genuinely correlated variables) is asking for trouble." (Steven S Skiena, "The Data Science Design Manual", 2017)

"The high generalization error in a neural network may be caused by several reasons. First, the data itself might have a lot of noise, in which case there is little one can do in order to improve accuracy. Second, neural networks are hard to train, and the large error might be caused by the poor convergence behavior of the algorithm. The error might also be caused by high bias, which is referred to as underfitting. Finally, overfitting (i.e., high variance) may cause a large part of the generalization error. In most cases, the error is a combination of more than one of these different factors." (Charu C Aggarwal, "Neural Networks and Deep Learning: A Textbook", 2018)

"[...] in the statistical world, what we see and measure around us can be considered as the sum of a systematic mathematical idealized form plus some random contribution that cannot yet be explained. This is the classic idea of the signal and the noise." (David Spiegelhalter, "The Art of Statistics: Learning from Data", 2019)

"Visualizations can remove the background noise from enormous sets of data so that only the most important points stand out to the intended audience. This is particularly important in the era of big data. The more data there is, the more chance for noise and outliers to interfere with the core concepts of the data set." (Kate Strachnyi, "ColorWise: A Data Storyteller’s Guide to the Intentional Use of Color", 2023)

🔭Data Science: Missing Data (Just the Quotes)

"Place little faith in an average or a graph or a trend when those important figures are missing."  (Darell Huff, "How to Lie with Statistics", 1954)

"Missing data values pose a particularly sticky problem for symbols. For instance, if the ray corresponding to a missing value is simply left off of a star symbol, the result will be almost indistinguishable from a minimum (i.e., an extreme) value. It may be better either (i) to impute a value, perhaps a median for that variable, or a fitted value from some regression on other variables, (ii) to indicate that the value is missing, possibly with a dashed line, or (iii) not to draw the symbol for a particular observation if any value is missing." (John M Chambers et al, "Graphical Methods for Data Analysis", 1983)

"The progress of science requires more than new data; it needs novel frameworks and contexts. And where do these fundamentally new views of the world arise? They are not simply discovered by pure observation; they require new modes of thought. And where can we find them, if old modes do not even include the right metaphors? The nature of true genius must lie in the elusive capacity to construct these new modes from apparent darkness. The basic chanciness and unpredictability of science must also reside in the inherent difficulty of such a task." (Stephen J Gould, "The Flamingo's Smile: Reflections in Natural History", 1985)

"We often think, naïvely, that missing data are the primary impediments to intellectual progress - just find the right facts and all problems will dissipate. But barriers are often deeper and more abstract in thought. We must have access to the right metaphor, not only to the requisite information. Revolutionary thinkers are not, primarily, gatherers of facts, but weavers of new intellectual structures." (Stephen J Gould, "The Flamingo's Smile: Reflections in Natural History", 1985)

"[...] as the planning process proceeds to a specific financial or marketing state, it is usually discovered that a considerable body of 'numbers' is missing, but needed numbers for which there has been no regular system of collection and reporting; numbers that must be collected outside the firm in some cases. This serendipity usually pays off in a much better management information system in the form of reports which will be collected and reviewed routinely." (William H. Franklin Jr., Financial Strategies, 1987)

"We have found that some of the hardest errors to detect by traditional methods are unsuspected gaps in the data collection (we usually discovered them serendipitously in the course of graphical checking)." (Peter Huber, "Huge data sets", Compstat ’94: Proceedings, 1994)

"Unfortunately, just collecting the data in one place and making it easily available isn’t enough. When operational data from transactions is loaded into the data warehouse, it often contains missing or inaccurate data. How good or bad the data is a function of the amount of input checking done in the application that generates the transaction. Unfortunately, many deployed applications are less than stellar when it comes to validating the inputs. To overcome this problem, the operational data must go through a 'cleansing' process, which takes care of missing or out-of-range values. If this cleansing step is not done before the data is loaded into the data warehouse, it will have to be performed repeatedly whenever that data is used in a data mining operation." (Joseph P Bigus,"Data Mining with Neural Networks: Solving business problems from application development to decision support", 1996)

"If you have only a small proportion of cases with missing data, you can simply throw out those cases for purposes of estimation; if you want to make predictions for cases with missing inputs, you don’t have the option of throwing those cases out." (Warren S Sarle, "Prediction with missing inputs", 1998)

"Every statistical analysis is an interpretation of the data, and missingness affects the interpretation. The challenge is that when the reasons for the missingness cannot be determined there is basically no way to make appropriate statistical adjustments. Sensitivity analyses are designed to model and explore a reasonable range of explanations in order to assess the robustness of the results." (Gerald van Belle, "Statistical Rules of Thumb", 2002)

"The best rule is: Don't have any missing data, Unfortunately, that is unrealistic. Therefore, plan for missing data and develop strategies to account for them. Do this before starting the study. The strategy should state explicitly how the type of missingness will be examined, how it will be handled, and how the sensitivity of the results to the missing data will be assessed." (Gerald van Belle, "Statistical Rules of Thumb", 2002)

"Statistics depend on collecting information. If questions go unasked, or if they are asked in ways that limit responses, or if measures count some cases but exclude others, information goes ungathered, and missing numbers result. Nevertheless, choices regarding which data to collect and how to go about collecting the information are inevitable." (Joel Best, "More Damned Lies and Statistics: How numbers confuse public issues", 2004)

"A sin of omission – leaving something out – is a strong one and not always recognized; itʼs hard to ask for something you donʼt know is missing. When looking into the data, even before it is graphed and charted, there is potential for abuse. Simply not having all the data or the correct data before telling your story can cause problems and unhappy endings." (Brian Suda, "A Practical Guide to Designing with Data", 2010)

"Having NUMBERSENSE means: (•) Not taking published data at face value; (•) Knowing which questions to ask; (•) Having a nose for doctored statistics. [...] NUMBERSENSE is that bit of skepticism, urge to probe, and desire to verify. It’s having the truffle hog’s nose to hunt the delicacies. Developing NUMBERSENSE takes training and patience. It is essential to know a few basic statistical concepts. Understanding the nature of means, medians, and percentile ranks is important. Breaking down ratios into components facilitates clear thinking. Ratios can also be interpreted as weighted averages, with those weights arranged by rules of inclusion and exclusion. Missing data must be carefully vetted, especially when they are substituted with statistical estimates. Blatant fraud, while difficult to detect, is often exposed by inconsistency." (Kaiser Fung, "Numbersense: How To Use Big Data To Your Advantage", 2013)

"Quality without science and research is absurd. You can't make inferences that something works when you have 60 percent missing data." (Peter Pronovost, "Safe Patients, Smart Hospitals", 2010)

"The only thing we know for sure about a missing data point is that it is not there, and there is nothing that the magic of statistics can do change that. The best that can be managed is to estimate the extent to which missing data have influenced the inferences we wish to draw." (Howard Wainer, "14 Conversations About Three Things", Journal of Educational and Behavioral Statistics Vol. 35(1, 2010)

"There are several key issues in the field of statistics that impact our analyses once data have been imported into a software program. These data issues are commonly referred to as the measurement scale of variables, restriction in the range of data, missing data values, outliers, linearity, and nonnormality." (Randall E Schumacker & Richard G Lomax, "A Beginner’s Guide to Structural Equation Modeling" 3rd Ed., 2010)

"Missing data is the blind spot of statisticians. If they are not paying full attention, they lose track of these little details. Even when they notice, many unwittingly sway things our way. Most ranking systems ignore missing values." (Kaiser Fung, "Numbersense: How To Use Big Data To Your Advantage", 2013)

"Accuracy and coherence are related concepts pertaining to data quality. Accuracy refers to the comprehensiveness or extent of missing data, performance of error edits, and other quality assurance strategies. Coherence is the degree to which data - item value and meaning are consistent over time and are comparable to similar variables from other routinely used data sources." (Aileen Rothbard, "Quality Issues in the Use of Administrative Data Records", 2015)

"How good the data quality is can be looked at both subjectively and objectively. The subjective component is based on the experience and needs of the stakeholders and can differ by who is being asked to judge it. For example, the data managers may see the data quality as excellent, but consumers may disagree. One way to assess it is to construct a survey for stakeholders and ask them about their perception of the data via a questionnaire. The other component of data quality is objective. Measuring the percentage of missing data elements, the degree of consistency between records, how quickly data can be retrieved on request, and the percentage of incorrect matches on identifiers (same identifier, different social security number, gender, date of birth) are some examples." (Aileen Rothbard, "Quality Issues in the Use of Administrative Data Records", 2015)

"When we find data quality issues due to valid data during data exploration, we should note these issues in a data quality plan for potential handling later in the project. The most common issues in this regard are missing values and outliers, which are both examples of noise in the data." (John D. Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, worked examples, and case studies", 2015)

"There are other problems with Big Data. In any large data set, there are bound to be inconsistencies, misclassifications, missing data - in other words, errors, blunders, and possibly lies. These problems with individual items occur in any data set, but they are often hidden in a large mass of numbers even when these numbers are generated out of computer interactions." (David S Salsburg, "Errors, Blunders, and Lies: How to Tell the Difference", 2017)

"Unless we’re collecting data ourselves, there’s a limit to how much we can do to combat the problem of missing data. But we can and should remember to ask who or what might be missing from the data we’re being told about. Some missing numbers are obvious […]. Other omissions show up only when we take a close look at the claim in question." (Tim Harford, "The Data Detective: Ten easy rules to make sense of statistics", 2020)

"[Making reasoned macro calls] starts with having the best and longest-time-series data you can find. You may have to take some risks in terms of the quality of data sources, but it amazes me how people are often more willing to act based on little or no data than to use data that is a challenge to assemble." (Robert J Shiller)

🔭Data Science: Signals (Just the Quotes)

"If statistical graphics, although born just yesterday, extends its reach every day, it is because it replaces long tables of numbers and it allows one not only to embrace at glance the series of phenomena, but also to signal the correspondences or anomalies, to find the causes, to identify the laws." (Émile Cheysson, cca. 1877)

"The term closed loop-learning process refers to the idea that one learns by determining what s desired and comparing what is actually taking place as measured at the process and feedback for comparison. The difference between what is desired and what is taking place provides an error indication which is used to develop a signal to the process being controlled." (Harold Chestnut, 1984) 

"Complexity is not an objective factor but a subjective one. Supersignals reduce complexity, collapsing a number of features into one. Consequently, complexity must be understood in terms of a specific individual and his or her supply of supersignals. We learn supersignals from experience, and our supply can differ greatly from another individual's. Therefore there can be no objective measure of complexity." (Dietrich Dorner, "The Logic of Failure: Recognizing and Avoiding Error in Complex Situations", 1989)

"An artificial neural network is an information-processing system that has certain performance characteristics in common with biological neural networks. Artificial neural networks have been developed as generalizations of mathematical models of human cognition or neural biology, based on the assumptions that: (1) Information processing occurs at many simple elements called neurons. (2) Signals are passed between neurons over connection links. (3) Each connection link has an associated weight, which, in a typical neural net, multiplies the signal transmitted. (4) Each neuron applies an activation function (usually nonlinear) to its net input (sum of weighted input signals) to determine its output signal." (Laurene Fausett, "Fundamentals of Neural Networks", 1994)

"Data are generally collected as a basis for action. However, unless potential signals are separated from probable noise, the actions taken may be totally inconsistent with the data. Thus, the proper use of data requires that you have simple and effective methods of analysis which will properly separate potential signals from probable noise." (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"No matter what the data, and no matter how the values are arranged and presented, you must always use some method of analysis to come up with an interpretation of the data.

While every data set contains noise, some data sets may contain signals. Therefore, before you can detect a signal within any given data set, you must first filter out the noise." (Donald J Wheeler," Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"While all data contain noise, some data contain signals. Before you can detect a signal, you must filter out the noise." (Donald J Wheeler, "Understanding Variation: The Key to Managing Chaos" 2nd Ed., 2000)

"The most familiar example of swarm intelligence is the human brain. Memory, perception and thought all arise out of the nett actions of billions of individual neurons. As we saw earlier, artificial neural networks (ANNs) try to mimic this idea. Signals from the outside world enter via an input layer of neurons. These pass the signal through a series of hidden layers, until the result emerges from an output layer. Each neuron modifies the signal in some simple way. It might, for instance, convert the inputs by plugging them into a polynomial, or some other simple function. Also, the network can learn by modifying the strength of the connections between neurons in different layers." (David G Green, "The Serendipity Machine: A voyage of discovery through the unexpected world of computers", 2004)

"Data analysis is not generally thought of as being simple or easy, but it can be. The first step is to understand that the purpose of data analysis is to separate any signals that may be contained within the data from the noise in the data. Once you have filtered out the noise, anything left over will be your potential signals. The rest is just details." (Donald J Wheeler," Myths About Data Analysis", International Lean & Six Sigma Conference, 2012)

"Finding patterns is easy in any kind of data-rich environment […] The key is in determining whether the patterns represent signal or noise." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don’t", 2012)

"The signal is the truth. The noise is what distracts us from the truth." (Nate Silver, "The Signal and the Noise: Why So Many Predictions Fail-but Some Don't", 2012)

"A signal is a useful message that resides in data. Data that isn’t useful is noise. […] When data is expressed visually, noise can exist not only as data that doesn’t inform but also as meaningless non-data elements of the display (e.g. irrelevant attributes, such as a third dimension of depth in bars, color variation that has no significance, and artificial light and shadow effects)." (Stephen Few, "Signal: Understanding What Matters in a World of Noise", 2015)

"To find signals in data, we must learn to reduce the noise - not just the noise that resides in the data, but also the noise that resides in us. It is nearly impossible for noisy minds to perceive anything but noise in data. […] Signals always point to something. In this sense, a signal is not a thing but a relationship. Data becomes useful knowledge of something that matters when it builds a bridge between a question and an answer. This connection is the signal." (Stephen Few, "Signal: Understanding What Matters in a World of Noise", 2015)

"Information theory leads to the quantification of the information content of the source, as denoted by entropy, the characterization of the information-bearing capacity of the communication channel, as related to its noise characteristics, and consequently the establishment of the relationship between the information content of the source and the capacity of the channel. In short, information theory provides a quantitative measure of the information contained in message signals and help determine the capacity of a communication system to transfer this information from source to sink over a noisy channel in a reliable fashion." (Ali Grami, "Information Theory", 2016)

🔭Data Science: Regression toward the Mean (Just the Quotes)

"Whenever we make any decision based on the expectation that matters will return to 'normal', we are employing the notion of regression to the mean." (Peter L Bernstein, "Against the Gods: The Remarkable Story of Risk", 1996)

"Regression to the mean occurs when the process produces results that are statistically independent or negatively correlated. With strong negative serial correlation, extremes are likely to be reversed each time (which would reinforce the instructors' error). In contrast, with strong positive dependence, extreme results are quite likely to be clustered together." (Dan Trietsch, "Statistical Quality Control : A loss minimization approach", 1998) 

"Unfortunately, people are poor intuitive scientists, generally failing to reason in accordance with the principles of scientific method. For example, people do not generate sufficient alternative explanations or consider enough rival hypotheses. People generally do not adequately control for confounding variables when they explore a novel environment. People’s judgments are strongly affected by the frame in which the information is presented, even when the objective information is unchanged. People suffer from overconfidence in their judgments (underestimating uncertainty), wishful thinking (assessing desired outcomes as more likely than undesired outcomes), and the illusion of control (believing one can predict or influence the outcome of random events). People violate basic rules of probability, do not understand basic statistical concepts such as regression to the mean, and do not update beliefs according to Bayes’ rule. Memory is distorted by hindsight, the availability and salience of examples, and the desirability of outcomes. And so on."  (John D Sterman, "Business Dynamics: Systems thinking and modeling for a complex world", 2000)

 "People often attribute meaning to phenomena governed only by a regression to the mean, the mathematical tendency for an extreme value of an at least partially chance-dependent quantity to be followed by a value closer to the average. Sports and business are certainly chancy enterprises and thus subject to regression. So is genetics to an extent, and so very tall parents can be expected to have offspring who are tall, but probably not as tall as they are. A similar tendency holds for the children of very short parents." (John A Paulos, "A Mathematician Plays the Stock Market", 2003)

"'Regression to the mean' […] says that, in any series of events where chance is involved, very good or bad performances, high or low scores, extreme events, etc. tend on the average, to be followed by more average performance or less extreme events. If we do extremely well, we're likely to do worse the next time, while if we do poorly, we're likely to do better the next time. But regression to the mean is not a natural law. Merely a statistical tendency. And it may take a long time before it happens." (Peter Bevelin, "Seeking Wisdom: From Darwin to Munger",  2003)

"Another aspect of representativeness that is misunderstood or ignored is the tendency of regression to the mean. Stochastic phenomena where the outcomes vary randomly around stable values (so-called stationary processes) exhibit the general tendency that extreme outcomes are more likely to be followed by an outcome closer to the mean or mode than by other extreme values in the same direction. For example, even a bright student will observe that her or his performance in a test following an especially outstanding outcome tends to be less brilliant. Similarly, extremely low or extremely high sales in a given period tend to be followed by sales that are closer to the stable mean or the stable trend." (Hans G Daellenbach & Donald C McNickle, "Management Science: Decision making through systems thinking", 2005)

"Behavioural research shows that we tend to use simplifying heuristics when making judgements about uncertain events. These are prone to biases and systematic errors, such as stereotyping, disregard of sample size, disregard for regression to the mean, deriving estimates based on the ease of retrieving instances of the event, anchoring to the initial frame, the gambler’s fallacy, and wishful thinking, which are all affected by our inability to consider more than a few aspects or dimensions of any phenomenon or situation at the same time." (Hans G Daellenbach & Donald C McNickle, "Management Science: Decision making through systems thinking", 2005)

"Concluding that the population is becoming more centralized by observing behavior at the extremes is called the 'Regression to the Mean' Fallacy. […] When looking for a change in a population, do not look only at the extremes; there you will always find a motion to the mean. Look at the entire population." (Charles Livingston & Paul Voakes, "Working with Numbers and Statistics: A handbook for journalists", 2005)

"'Regression to the mean' describes a natural phenomenon whereby, after a short period of success, things tend to return to normal immediately afterwards. This notion applies particularly to random events." (Alan Graham, "Developing Thinking in Statistics", 2006)

"regression to the mean: The fact that unexpectedly high or low numbers from the mean are an exception and are usually followed by numbers that are closer to the mean. Over the long haul, we tend to get relatively more numbers that are near the mean compared to numbers that are far from the mean." (Hari Singh, "Framed! Solve an Intriguing Mystery and Master How to Make Smart Choices", 2006)

 "A naive interpretation of regression to the mean is that heights, or baseball records, or other variable phenomena necessarily become more and more 'average' over time. This view is mistaken because it ignores the error in the regression predicting y from x. For any data point xi, the point prediction for its yi will be regressed toward the mean, but the actual yi that is observed will not be exactly where it is predicted. Some points end up falling closer to the mean and some fall further." (Andrew Gelman & Jennifer Hill, "Data Analysis Using Regression and Multilevel/Hierarchical Models", 2007)

"Regression toward the mean. That is, in any series of random events an extraordinary event is most likely to be followed, due purely to chance, by a more ordinary one." (Leonard Mlodinow, "The Drunkard’s Walk: How Randomness Rules Our Lives", 2008)

"Regression does not describe changes in ability that happen as time passes […]. Regression is caused by performances fluctuating about ability, so that performances far from the mean reflect abilities that are closer to the mean." (Gary Smith, "Standard Deviations", 2014)

"We encounter regression in many contexts - pretty much whenever we see an imperfect measure of what we are trying to measure. Standardized tests are obviously an imperfect measure of ability. [...] Each experimental score is an imperfect measure of “ability,” the benefits from the layout. To the extent there is randomness in this experiment - and there surely is - the prospective benefits from the layout that has the highest score are probably closer to the mean than was the score." (Gary Smith, "Standard Deviations", 2014)

"When a trait, such as academic or athletic ability, is measured imperfectly, the observed differences in performance exaggerate the actual differences in ability. Those who perform the best are probably not as far above average as they seem. Nor are those who perform the worst as far below average as they seem. Their subsequent performances will consequently regress to the mean." (Gary Smith, "Standard Deviations", 2014)

"The term shrinkage is used in regression modeling to denote two ideas. The first meaning relates to the slope of a calibration plot, which is a plot of observed responses against predicted responses. When a dataset is used to fit the model parameters as well as to obtain the calibration plot, the usual estimation process will force the slope of observed versus predicted values to be one. When, however, parameter estimates are derived from one dataset and then applied to predict outcomes on an independent dataset, overfitting will cause the slope of the calibration plot (i.e., the shrinkage factor ) to be less than one, a result of regression to the mean. Typically, low predictions will be too low and high predictions too high. Predictions near the mean predicted value will usually be quite accurate. The second meaning of shrinkage is a statistical estimation method that preshrinks regression coefficients towards zero so that the calibration plot for new data will not need shrinkage as its calibration slope will be one." (Frank E. Harrell Jr., "Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis" 2nd Ed, 2015)

"Often when people relate essentially the same variable in two different groups, or at two different times, they see this same phenomenon - the tendency of the response variable to be closer to the mean than the predicted value. Unfortunately, people try to interpret this by thinking that the performance of those far from the mean is deteriorating, but it’s just a mathematical fact about the correlation. So, today we try to be less judgmental about this phenomenon and we call it regression to the mean. We managed to get rid of the term 'mediocrity', but the name regression stuck as a name for the whole least squares fitting procedure - and that’s where we get the term regression line." (Richard D De Veaux et al, "Stats: Data and Models", 2016)

"Regression toward the mean is pervasive. In sports, excellent performance tends to be followed by good, but less outstanding, performance. [...] By contrast, the good news about regression toward the mean is that very poor performance tends to be followed by improved performance. If you got the worst score in your statistics class on the first exam, you probably did not do so poorly on the second exam (but you were probably still below the mean)." (Alan Agresti et al, Statistics: The Art and Science of Learning from Data" 4th Ed., 2018)

🔭Data Science: Decision Trees (Just the Quotes)

"A good way to evaluate a model is to look at a visual representation of it. After all, what is easier to understand - a table full of mathematical relationships or a graphic displaying a decision tree with all of its splits and branches?" (Seth Paul et al. "Preparing and Mining Data with Microsoft SQL Server 2000 and Analysis", 2002)

"When generating trees, it is usually optimal to grow a larger tree than is justifiable and then prune it back. The main reason this works well is that stop splitting rules do not look far enough forward. That is, stop splitting rules tend to underfit, meaning that even if a rule stops at a split for which the next candidate splits give little improvement, it may be that splitting them one layer further will give a large improvement in accuracy." (Bertrand Clarke et al, "Principles and Theory for Data Mining and Machine Learning", 2009)

"A predictive model overfits the training set when at least some of the predictions it returns are based on spurious patterns present in the training data used to induce the model. Overfitting happens for a number of reasons, including sampling variance and noise in the training set. The problem of overfitting can affect any machine learning algorithm; however, the fact that decision tree induction algorithms work by recursively splitting the training data means that they have a natural tendency to segregate noisy instances and to create leaf nodes around these instances. Consequently, decision trees overfit by splitting the data on irrelevant features that only appear relevant due to noise or sampling variance in the training data. The likelihood of overfitting occurring increases as a tree gets deeper because the resulting predictions are based on smaller and smaller subsets as the dataset is partitioned after each feature test in the path." (John D Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, Worked Examples, and Case Studies", 2015)

"Decision trees are also discriminative models. Decision trees are induced by recursively partitioning the feature space into regions belonging to the different classes, and consequently they define a decision boundary by aggregating the neighboring regions belonging to the same class. Decision tree model ensembles based on bagging and boosting are also discriminative models." (John D Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, Worked Examples, and Case Studies", 2015)

"Decision trees are also considered nonparametric models. The reason for this is that when we train a decision tree from data, we do not assume a fixed set of parameters prior to training that define the tree. Instead, the tree branching and the depth of the tree are related to the complexity of the dataset it is trained on. If new instances were added to the dataset and we rebuilt the tree, it is likely that we would end up with a (potentially very) different tree." (John D Kelleher et al, "Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms, Worked Examples, and Case Studies", 2015)

"Decision trees are considered a good predictive model to start with, and have many advantages. Interpretability, variable selection, variable interaction, and the flexibility to choose the level of complexity for a decision tree all come into play." (Ralph Winters, "Practical Predictive Analytics", 2017)

"Decision trees are important for a few reasons. First, they can both classify and regress. It requires literally one line of code to switch between the two models just described, from a classification to a regression. Second, they are able to determine and share the feature importance of a given training set." (Russell Jurney, "Agile Data Science 2.0: Building Full-Stack Data Analytics Applications with Spark", 2017)

 "The main reason why pruning tends to improve classification performance on future examples is that the removal of low-level tests, which have poor statistical support, usually reduces the danger of overfitting. This, however, works only up to a certain point. If overdone, a very high extent of pruning can (in the extreme) result in the decision being replaced with a single leaf labeled with the majority class." (Miroslav Kubat," An Introduction to Machine Learning" 2nd Ed., 2017)

"From a typical training set, many alternative decision trees can be created. As a rule, smaller trees are to be preferred, their main advantages being interpretability, removal of irrelevant and redundant attributes, and lower danger of overfitting noisy training data." (Miroslav Kubat, "An Introduction to Machine Learning" 2nd Ed., 2017)

"An advantage of random forests is that it works with both regression and classification trees so it can be used with targets whose role is binary, nominal, or interval. They are also less prone to overfitting than a single decision tree model. A disadvantage of a random forest is that they generally require more trees to improve their accuracy. This can result in increased run times, particularly when using very large data sets." (Richard V McCarthy et al, "Applying Predictive Analytics: Finding Value in Data", 2019)

"Decision trees show the breakdown of the data by one variable then another in a very intuitive way, though they are generally just diagrams that don’t actually encode data visually." (Robert Grant, "Data Visualization: Charts, Maps and Interactive Graphics", 2019)

"Random forests are essentially an ensemble of trees. They use many short trees, fitted to multiple samples of the data, and the predictions are averaged for each observation. This helps to get around a problem that trees, and many other machine learning techniques, are not guaranteed to find optimal models, in the way that linear regression is. They do a very challenging job of fitting non-linear predictions over many variables, even sometimes when there are more variables than there are observations. To do that, they have to employ 'greedy algorithms', which find a reasonably good model but not necessarily the very best model possible." (Robert Grant, "Data Visualization: Charts, Maps and Interactive Graphics", 2019)

🔭Data Science: Evidence (Just the Quotes)

"Rule 1. Original data should be presented in a way that will preserve the evidence in the original data for all the predictions assumed to be useful." (Walter A Shewhart, "Economic Control of Quality of Manufactured Product", 1931)

"Rule 2. Any summary of a distribution of numbers in terms of symmetric functions should not give an objective degree of belief in any one of the inferences or predictions to be made therefrom that would cause human action significantly different from what this action would be if the original distributions had been taken as evidence." (Walter A Shewhart, "Economic Control of Quality of Manufactured Product", 1931)

"[...] there is evidence that significance tests have been a genuine block to achieving [...] knowledge." (Denton E Morrison & Ramon E Henkel, "Significance tests reconsidered", The American Sociologist 4, 1969)

"Confidence intervals give a feeling of the uncertainty of experimental evidence, and (very important) give it in the same units [...] as the original observations." (Mary G Natrella, "The relation between confidence intervals and tests of significance", American Statistician 14, 1960)

"The null-hypothesis significance test treats ‘acceptance’ or ‘rejection’ of a hypothesis as though these were decisions one makes. But a hypothesis is not something, like a piece of pie offered for dessert, which can be accepted or rejected by a voluntary physical action. Acceptance or rejection of a hypothesis is a cognitive process, a degree of believing or disbelieving which, if rational, is not a matter of choice but determined solely by how likely it is, given the evidence, that the hypothesis is true." (William W Rozeboom, "The fallacy of the null–hypothesis significance test", Psychological Bulletin 57, 1960)

"Scientific discovery, or the formulation of scientific theory, starts in with the unvarnished and unembroidered evidence of the senses. It starts with simple observation - simple, unbiased, unprejudiced, naive, or innocent observation - and out of this sensory evidence, embodied in the form of simple propositions or declarations of fact, generalizations will grow up and take shape, almost as if some process of crystallization or condensation were taking place. Out of a disorderly array of facts, an orderly theory, an orderly general statement, will somehow emerge." (Sir Peter B Medawar, "Is the Scientific Paper Fraudulent?", The Saturday Review, 1964)

"We have overwhelming evidence that available information plus analysis does not lead to knowledge. The management science team can properly analyse a situation and present recommendations to the manager, but no change occurs. The situation is so familiar to those of us who try to practice management science that I hardly need to describe the cases." (C West Churchman, "Managerial acceptance of scientific recommendations", California Management Review Vol 7, 1964)

"Science consists simply of the formulation and testing of hypotheses based on observational evidence; experiments are important where applicable, but their function is merely to simplify observation by imposing controlled conditions." (Henry L Batten, "Evolution of the Earth", 1971)

"Statistics is a body of methods and theory applied to numerical evidence in making decisions in the face of uncertainty." (Lawrence Lapin, "Statistics for Modern Business Decisions", 1973)

"The language of association and prediction is probably most often used because the evidence seems insufficient to justify a direct causal statement. A better practice is to state the causal hypothesis and then to present the evidence along with an assessment with respect to the causal hypothesis - instead of letting the quality of the data determine the language of the explanation." (Edward R Tufte, "Data Analysis for Politics and Policy", 1974)

"All interpretations made by a scientist are hypotheses, and all hypotheses are tentative. They must forever be tested and they must be revised if found to be unsatisfactory. Hence, a change of mind in a scientist, and particularly in a great scientist, is not only not a sign of weakness but rather evidence for continuing attention to the respective problem and an ability to test the hypothesis again and again." (Ernst Mayr, "The Growth of Biological Thought: Diversity, Evolution and Inheritance", 1982)

"It has been widely felt, probably for thirty years and more, that significance tests are overemphasized and often misused and that more emphasis should be put on estimation and prediction. While such a shift of emphasis does seem to be occurring, for example in medical statistics, the continued very extensive use of significance tests is on the one hand alarming and on the other evidence that they are aimed, even if imperfectly, at some widely felt need." (David R Cox, "Some general aspects of the theory of statistics", International Statistical Review 54, 1986)

"Like a detective, a data analyst will experience many dead ends, retrace his steps, and explore many alternatives before settling on a single description of the evidence in front of him." (David Lubinsky & Daryl Pregibon , "Data analysis as search", Journal of Econometrics Vol. 38 (1–2), 1988)

"Subjective probability, also known as Bayesian statistics, pushes Bayes' theorem further by applying it to statements of the type described as 'unscientific' in the frequency definition. The probability of a theory (e.g. that it will rain tomorrow or that parity is not violated) is considered to be a subjective 'degree of belief - it can perhaps be measured by seeing what odds the person concerned will offer as a bet. Subsequent experimental evidence then modifies the initial degree of belief, making it stronger or weaker according to whether the results agree or disagree with the predictions of the theory in question." (Roger J Barlow, "Statistics: A guide to the use of statistical methods in the physical sciences", 1989)

"Probability theory is an ideal tool for formalizing uncertainty in situations where class frequencies are known or where evidence is based on outcomes of a sufficiently long series of independent random experiments. Possibility theory, on the other hand, is ideal for formalizing incomplete information expressed in terms of fuzzy propositions." (George Klir, "Fuzzy sets and fuzzy logic", 1995)

"[…] the simplest hypothesis proposed as an explanation of phenomena is more likely to be the true one than is any other available hypothesis, that its predictions are more likely to be true than those of any other available hypothesis, and that it is an ultimate a priori epistemic principle that simplicity is evidence for truth." (Richard Swinburne, "Simplicity as Evidence for Truth", 1997)

"When significance tests are used and a null hypothesis is not rejected, a major problem often arises - namely, the result may be interpreted, without a logical basis, as providing evidence for the null hypothesis." (David F Parkhurst, "Statistical Significance Tests: Equivalence and Reverse Tests Should Reduce Misinterpretation", BioScience Vol. 51 (12), 2001)

"One cautious approach is represented by Bernoulli’s more conservative outlook. If there are very strong reasons for believing that an observation has suffered an accident that made the value in the data-file thoroughly untrustworthy, then reject it; in the absence of clear evidence that an observation, identified by formal rule as an outlier, is unacceptable then retain it unless there is lack of trust that the laboratory obtaining it is conscientiously operated by able persons who have [...] taken every care.'" (David Finney, "Calibration Guidelines Challenge Outlier Practices", The American Statistician Vol 60 (4), 2006)

"Scholars feel the need to present tables of model parameters in academic articles (perhaps just as evidence that they ran the analysis they claimed to have run), but these tables are rarely interpreted other than for their sign and statistical significance. Most of the numbers in these tables are never even discussed in the text. From the perspective of the applied data analyst, R packages without procedures to compute quantities of scientific interest are woefully incomplete. A better approach focuses on quantities of direct scientific interest rather than uninterpretable model parameters. [...] For each quantity of interest, the user needs some summary that includes a point estimate and a measure of uncertainty such as a standard error, confidence interval, or a distribution. The methods of calculating these differ greatly across theories of inference and methods of analysis. However, from the user’s perspective, the result is almost always the same: the point estimate and uncertainty of some quantity of interest." (Kousuke Imai et al, "Toward a Common Framework for Statistical Analysis and Development", Journal of Computational and Graphical Statistics vol. 17, 2008)

"Data analysis is careful thinking about evidence." (Michael Milton, "Head First Data Analysis", 2009)

"Data clusters are everywhere, even in random data. Someone who looks for an explanation will inevitably find one, but a theory that fits a data cluster is not persuasive evidence. The found explanation needs to make sense and it needs to be tested with uncontaminated data." (Gary Smith, "Standard Deviations", 2014)

"In general, when building statistical models, we must not forget that the aim is to understand something about the real world. Or predict, choose an action, make a decision, summarize evidence, and so on, but always about the real world, not an abstract mathematical world: our models are not the reality - a point well made by George Box in his oft-cited remark that "all models are wrong, but some are useful". (David Hand, "Wonderful examples, but let's not close our eyes", Statistical Science 29, 2014)

"The dialectical interplay of experiment and theory is a key driving force of modern science. Experimental data do only have meaning in the light of a particular model or at least a theoretical background. Reversely theoretical considerations may be logically consistent as well as intellectually elegant: Without experimental evidence they are a mere exercise of thought no matter how difficult they are. Data analysis is a connector between experiment and theory: Its techniques advise possibilities of model extraction as well as model testing with experimental data." (Achim Zielesny, "From Curve Fitting to Machine Learning" 2nd Ed., 2016)

"In terms of characteristics, a data scientist has an inquisitive mind and is prepared to explore and ask questions, examine assumptions and analyse processes, test hypotheses and try out solutions and, based on evidence, communicate informed conclusions, recommendations and caveats to stakeholders and decision makers." (Jesús Rogel-Salazar, "Data Science and Analytics with Python", 2017)

"With the growing availability of massive data sets and user-friendly analysis software, it might be thought that there is less need for training in statistical methods. This would be naïve in the extreme. Far from freeing us from the need for statistical skills, bigger data and the rise in the number and complexity of scientific studies makes it even more difficult to draw appropriate conclusions. More data means that we need to be even more aware of what the evidence is actually worth." (David Spiegelhalter, "The Art of Statistics: Learning from Data", 2019)

"The general principles of starting with a well-defined question, engaging in careful observation, and then formulating hypotheses and assessing the strength of evidence for and against them became known as the scientific method." (Michael Friendly & Howard Wainer, "A History of Data Visualization and Graphic Communication", 2021)

"Absence of evidence is not evidence of absence." (Martin Rees)

"The deepest sin of the human mind is to believe things without evidence." (Thomas H Huxley)