DOCSLIB.ORG

Inferential Statistics

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Inferential Statistics STATISTICS Milo Schield Ó Spring Day, 2001 SL2000T1 SL2000T1 TABLE OF CONTENTS DESCRIPTIVE STATISTICS Part I: Fundamentals 1. Causality and Statistics 2. Describing Count-Based Data 3. Comparing Count-Based Data 4. Interpreting Count-Based Data 5. Reading and Interpreting Measurements Part II: Graphs and Models 6. Graphs 7. Linear Models – Single Factor 8. Linear Models – Multiple Factors 9. Non-linear Models INFERENTIAL STATISTICS Part III: Probability and Inferential Statistics 10. Obtaining Statistics: Samples, Surveys and Experiments 11. Statistical Expectation, Chance and Error 12. Statistical Confidence in Estimates 13. Statistical Significance in Judgements Part IV: Statistical Inference and Decision making 14. Interpreting Statistical Arguments involving Chance Page a DEDICATION to Florence Nightingale, Ayn Rand and Julian Simon MISSION To help students read and interpret statistics as evidence in arguments Page b Page c Introduction TO STUDENTS Statistics are different; Statistics is important. 1 Statistics is not just arithmetic. Arithmetic deals with certainties: 2+2 = 4. Statistics deals with uncertainties: the chance of two heads in flipping a fair coin twice is 25%. 2 Statistics are not just numbers. Numbers just are: 2+2 = 4. Statistics describe reality: for US teens age 15-17, the birth rate for Asians is half of the birth rate for American Indians. 3 Statistics is essential to understanding the sciences. Both the social sciences (psychology, sociology, economics and political science) and the physical sciences (physics, chemistry, biology) rely on statistics for explanations and predictions. 4 Statistics is a language just like accounting, finance or economics. All four disciplines involve measurements and use common words in technical ways. But, in terms of measurements, statistics is the most gen- eral; statistics is the language of data. Statistical literacy is different from traditional statistics. · Traditional statistics focuses on role of chance in confidence inter- vals and hypothesis tests. In traditional statistics, there is typically an answer is either right or wrong and it can be proved. · Statistical literacy studies the use of statistics as evidence in argu- ments. Statistical literacy focuses on the role of factors or models as ex- planations or as predictors. In statistical literacy, factors or models are either weaker or stronger in their ability to explain or predict. Science of Statistical literacy is a science of method – as are mathematics & statis- Method tics, logic and critical thinking. Sciences of method are fundamental to human thought. They can be classified by their content and by their method of reasoning. Table 1 --------- METHOD OF REASONING ------ FORMAL/SYMBOLIC INFORMAL/PRACTICAL CONTENT (Deductive) (Inductive) WORDS Logic Critical Thinking NUMBERS Mathematics; Probability & Statistical Literacy Traditional Statistics Page i Milo Schield Introduction Methods of · Logic, mathematics and traditional statistics focus on symbolic reason- reasoning ing using deductive arguments. If a deductive argument is valid, the answer is certain to be true given the truth of the premises. Although lim- ited in scope, this formal certainty is very valuable. · Critical thinking and statistical literacy focus on practical reasoning us- ing inductive arguments. These inductive arguments are on a spectrum from very weak to very strong. Although practical reasoning lacks formal validity, it is much broader in scope; practical reasoning is the common sense reasoning we do everyday. Content is more Most students view the distinction between words and numbers (rows) important than as more important than the distinction between the methods of reason- method ing (columns). The row-distinction is easier to recognize (numbers versus words). The row-distinction explains how these four courses are related to academic departments. Critical thinking and logic are part of Philosophy, statistics and probability are typically part of Mathematics. Method is This book argues that method is more important than content; the dis- more funda- tinction between symbolic reasoning and practical reasoning (columns) is mental than more fundamental than the distinction between words and numbers content (rows). This book views statistical literacy as being closer to critical thinking than to mathematics. As human beings, inductive reasoning is our primary method of thought. We are not omniscient; we need to choose know- ing our choice may be in error. We need to act knowing out actions may have consequences that were unforeseen and perhaps unforeseeable. Thus, our method of thinking is primary; the content of our thinking is secondary. Challenge This emphasis on statistical literacy may be challenging for those who · prefer the certainty of mathematics or logic, or · have difficulty with the English language. Both will be challenged to read and interpret statistics as evidence. This book is different -- radically different! This book is dedicated to help- ing you learn to think more effectively by using statistics as evidence. This is not a course where students memorize, regurgitate and forget mathe- matical statistics and then say, "I just didn't get it." This course focuses on the thinking we do every day. The principles learned in this book can help you think with more clarity and certainty. If you can master these principles, you will have an insight that many people lack. And with this insight you can work more effectively toward achieving wisdom in all areas of your life. For more on the author, see his web site at www.augsburg.edu/ppages/~schield or contact him at [email protected]. Page ii Milo Schield Introduction THE BOOK This book has features not found in any other statistics book to date: · an over-arching focus on statistical literacy · a focus on arguing about causality and the effects of control. · Ch. 3: Reading count data in tables · Ch. 6 and 9: Reading and interpreting graphs. · Ch. 16: Interpreting confidence and significance for decisions. Since this book has a very non-standard approach, it will be liked by some and disliked by others This book is divided into two sections: Descriptive Statistics and Inferential Statistics. Inferential statistics focuses entirely on the variability due to chance. Descriptive statistics ignores this topic. Descriptive statistics - Part 1 deals with fundamentals of practical reasoning (1) and with reading and interpreting the data found in newspapers, essays and books (2, 3, 4 and 5). In each case, the focus is on communication: de- scribing the data so others can understand it. Descriptive statistics - Part 2 deals with graphs (6) and models (7, 8 and 9). The focus is using the graph or model to explain or predict. Inferential statistics – Part 3 deals with statistical studies (10), sampling and probability (11), confidence intervals (11) and hypothesis tests (12). This part is the main topic in most statistics texts. Inferential statistics – Part 4 deals with interpreting confidence intervals and hypothesis tests in terms of decision-making (14). Other books Other books having a similar focus on practical reasoning include: · Critical thinking: "The Art of Reasoning" by David Kelley "Logical Analysis: A New Approach" by Richard Connell · Language: "Twice As Less" by Eleanor Orr · Economics: "Hidden Order" by David Friedman · Law: "Prove It With Figures" by Hans Zeisel and David Kaye · Journalism: "News and Numbers" by Victor Cohn Benefit The primary benefit of study this book is to help readers improve their think- ing – their ability to reason about arguments – and thus to make better deci- sions in pursuit of their own happiness and well being. Page iii Milo Schield Introduction THE AUTHOR Author Milo Schield, Associate Professor at Augsburg College, 2211 Riverside Avenue, Minneapolis, MN 55454 since 1985. He can be contacted at [email protected]. Author’s Dr. Schield has a Bachelors degree in physics from Iowa State University, a academic and Masters degree in physics from the University of Illinois, and a Ph.D. in space intellectual physics from Rice University. He has done post-graduate work at the Uni- credentials versity of Iowa in Economics and at the University of Minnesota in Insurance and in Business, Government and Society. He has studied philosophy on an on-going basis. Author’s Dr. Schield has been a Senior Consultant with a national CPA firm for 2 professional years, a Senior Operations Research Analyst for a large property-casualty credentials insurance company for 8 years, and a co-founder and President of a small computer business for 5 years. He earned a certificate in Managerial Ac- counting (CMA). Author’s Dr. Schield has taught for over 15 years. He has taught at the University of teaching Iowa, National College, the University of St. Thomas and Augsburg College. credentials He has taught traditional undergraduates, adult undergraduates and graduate students. He has taught a variety of subjects including accounting, finance, microeconomics, critical thinking, and statistics. Author’s Doctor Schield has been a visiting scholar at the Royal Statistical Societies’ statistical Centre for Statistical Education at the University of Nottingham in Nottingham credentials England. Dr. Schield has given papers on statistical education and statistical literacy at numerous conferences including ICOTS-5 (Singapore), ICME-6 (Tokyo), ASA, MSMESB, AMATYC and APDU. He has given talks at various colleges in England, Scotland and Wales, China and Australia.
Load more

Recommended publications
  • 1.2. INDUCTIVE REASONING Much Mathematical Discovery Starts with Inductive Reasoning – the Process of Reaching General Conclus
    1.2. INDUCTIVE REASONING Much mathematical discovery starts with inductive reasoning – the process of reaching general conclusions, called conjectures, through the examination of particular cases and the recognition of patterns. These conjectures are then more formally proved using deductive methods, which will be discussed in the next section. Below we look at three examples that use inductive reasoning. Number Patterns Suppose you had to predict the sixth number in the following sequence: 1, -3, 6, -10, 15, ? How would you proceed with such a question? The trick, it seems, is to discern a specific pattern in the given sequence of numbers. This kind of approach is a classic example of inductive reasoning. By identifying a rule that generates all five numbers in this sequence, your hope is to establish a pattern and be in a position to predict the sixth number with confidence. So can you do it? Try to work out this number before continuing. One fact is immediately clear: the numbers alternate sign from positive to negative. We thus expect the answer to be a negative number since it follows 15, a positive number. On closer inspection, we also realize that the difference between the magnitude (or absolute value) of successive numbers increases by 1 each time: 3 – 1 = 2 6 – 3 = 3 10 – 6 = 4 15 – 10 = 5 … We have then found the rule we sought for generating the next number in this sequence. The sixth number should be -21 since the difference between 15 and 21 is 6 and we had already determined that the answer should be negative.
    [Show full text]
  • There Is No Pure Empirical Reasoning
    There Is No Pure Empirical Reasoning 1. Empiricism and the Question of Empirical Reasons Empiricism may be defined as the view there is no a priori justification for any synthetic claim. Critics object that empiricism cannot account for all the kinds of knowledge we seem to possess, such as moral knowledge, metaphysical knowledge, mathematical knowledge, and modal knowledge.1 In some cases, empiricists try to account for these types of knowledge; in other cases, they shrug off the objections, happily concluding, for example, that there is no moral knowledge, or that there is no metaphysical knowledge.2 But empiricism cannot shrug off just any type of knowledge; to be minimally plausible, empiricism must, for example, at least be able to account for paradigm instances of empirical knowledge, including especially scientific knowledge. Empirical knowledge can be divided into three categories: (a) knowledge by direct observation; (b) knowledge that is deductively inferred from observations; and (c) knowledge that is non-deductively inferred from observations, including knowledge arrived at by induction and inference to the best explanation. Category (c) includes all scientific knowledge. This category is of particular import to empiricists, many of whom take scientific knowledge as a sort of paradigm for knowledge in general; indeed, this forms a central source of motivation for empiricism.3 Thus, if there is any kind of knowledge that empiricists need to be able to account for, it is knowledge of type (c). I use the term “empirical reasoning” to refer to the reasoning involved in acquiring this type of knowledge – that is, to any instance of reasoning in which (i) the premises are justified directly by observation, (ii) the reasoning is non- deductive, and (iii) the reasoning provides adequate justification for the conclusion.
    [Show full text]
  • A Philosophical Treatise on the Connection of Scientific Reasoning
    mathematics Review A Philosophical Treatise on the Connection of Scientific Reasoning with Fuzzy Logic Evangelos Athanassopoulos 1 and Michael Gr. Voskoglou 2,* 1 Independent Researcher, Giannakopoulou 39, 27300 Gastouni, Greece; [email protected] 2 Department of Applied Mathematics, Graduate Technological Educational Institute of Western Greece, 22334 Patras, Greece * Correspondence: [email protected] Received: 4 May 2020; Accepted: 19 May 2020; Published:1 June 2020 Abstract: The present article studies the connection of scientific reasoning with fuzzy logic. Induction and deduction are the two main types of human reasoning. Although deduction is the basis of the scientific method, almost all the scientific progress (with pure mathematics being probably the unique exception) has its roots to inductive reasoning. Fuzzy logic gives to the disdainful by the classical/bivalent logic induction its proper place and importance as a fundamental component of the scientific reasoning. The error of induction is transferred to deductive reasoning through its premises. Consequently, although deduction is always a valid process, it is not an infallible method. Thus, there is a need of quantifying the degree of truth not only of the inductive, but also of the deductive arguments. In the former case, probability and statistics and of course fuzzy logic in cases of imprecision are the tools available for this purpose. In the latter case, the Bayesian probabilities play a dominant role. As many specialists argue nowadays, the whole science could be viewed as a Bayesian process. A timely example, concerning the validity of the viruses’ tests, is presented, illustrating the importance of the Bayesian processes for scientific reasoning.
    [Show full text]
  • Uses for Census Data in Market Research
    Uses for Census Data in Market Research MRS Presentation 4 July 2011 Andrew Zelin, Director, Sampling & RMC, Ipsos MORI Why Census Data is so critical to us Use of Census Data underlies almost all of our quantitative work activities and projects Needed to ensure that: – Samples are balanced and representative – ie Sampling; – Results derived from samples reflect that of target population – ie Weighting; – Understanding how demographic and geographic factors affect the findings from surveys – ie Analysis. How would we do our jobs without the use of Census Data? Our work withoutCensusData Our work Regional Assembly What I will cover Introduction, overview and why Census Data is so critical; Census data for sampling; Census data for survey weighting; Census data for statistical analysis; Use of SARs; The future (eg use of “hypercubes”); Our work without Census Data / use of alternatives; Conclusions Census Data for Sampling For Random (pre-selected) Samples: – Need to define stratum sample sizes and sampling fractions; – Need to define cluster size, esp. if PPS – (Need to define stratum / cluster) membership; For any type of Sample: – To balance samples across demographics that cannot be included as quota controls or stratification variables – To determine number of sample points by geography – To create accurate booster samples (eg young people / ethnic groups) Census Data for Sampling For Quota (non-probability) Samples: – To set appropriate quotas (based on demographics); Data are used here at very localised level – ie Census Outputs Census Data for Sampling Census Data for Survey Weighting To ensure that the sample profile balances the population profile Census Data are needed to tell you what the population profile is ….and hence provide accurate weights ….and hence provide an accurate measure of the design effect and the true level of statistical reliability / confidence intervals on your survey measures But also pre-weighting, for interviewer field dept.
    [Show full text]
  • Argument, Structure, and Credibility in Public Health Writing Donald Halstead Instructor and Director of Writing Programs Harvard TH Chan School of Public Heath
    Argument, Structure, and Credibility in Public Health Writing Donald Halstead Instructor and Director of Writing Programs Harvard TH Chan School of Public Heath Some of the most important questions we face in public health include what policies we should follow, which programs and research should we fund, how and where we should intervene, and what our priorities should be in the face of overwhelming needs and scarce resources. These questions, like many others, are best decided on the basis of arguments, a word that has its roots in the Latin arguere, to make clear. Yet arguments themselves vary greatly in terms of their strength, accuracy, and validity. Furthermore, public health experts often disagree on matters of research, policy and practice, citing conflicting evidence and arriving at conflicting conclusions. As a result, critical readers, such as researchers, policymakers, journal editors and reviewers, approach arguments with considerable skepticism. After all, they are not going to change their programs, priorities, practices, research agendas or budgets without very solid evidence that it is necessary, feasible, and beneficial. This raises an important challenge for public health writers: How can you best make your case, in the face of so much conflicting evidence? To illustrate, let’s assume that you’ve been researching mother-to-child transmission (MTCT) of HIV in a sub-Saharan African country and have concluded that (claim) the country’s maternal programs for HIV counseling and infant nutrition should be integrated because (reasons) this would be more efficient in decreasing MTCT, improving child nutrition, and using scant resources efficiently. The evidence to back up your claim might consist of original research you have conducted that included program assessments and interviews with health workers in the field, your assessment of the other relevant research, the experiences of programs in other countries, and new WHO guidelines.
    [Show full text]
  • CORRELATION COEFFICIENTS Ice Cream and Crimedistribute Difficulty Scale ☺ ☺ (Moderately Hard)Or
    5 COMPUTING CORRELATION COEFFICIENTS Ice Cream and Crimedistribute Difficulty Scale ☺ ☺ (moderately hard)or WHAT YOU WILLpost, LEARN IN THIS CHAPTER • Understanding what correlations are and how they work • Computing a simple correlation coefficient • Interpretingcopy, the value of the correlation coefficient • Understanding what other types of correlations exist and when they notshould be used Do WHAT ARE CORRELATIONS ALL ABOUT? Measures of central tendency and measures of variability are not the only descrip- tive statistics that we are interested in using to get a picture of what a set of scores 76 Copyright ©2020 by SAGE Publications, Inc. This work may not be reproduced or distributed in any form or by any means without express written permission of the publisher. Chapter 5 ■ Computing Correlation Coefficients 77 looks like. You have already learned that knowing the values of the one most repre- sentative score (central tendency) and a measure of spread or dispersion (variability) is critical for describing the characteristics of a distribution. However, sometimes we are as interested in the relationship between variables—or, to be more precise, how the value of one variable changes when the value of another variable changes. The way we express this interest is through the computation of a simple correlation coefficient. For example, what’s the relationship between age and strength? Income and years of education? Memory skills and amount of drug use? Your political attitudes and the attitudes of your parents? A correlation coefficient is a numerical index that reflects the relationship or asso- ciation between two variables. The value of this descriptive statistic ranges between −1.00 and +1.00.
    [Show full text]
  • Psychology 205, Revelle, Fall 2014 Research Methods in Psychology Mid-Term
    Psychology 205, Revelle, Fall 2014 Research Methods in Psychology Mid-Term Name: ________________________________ 1. (2 points) What is the primary advantage of using the median instead of the mean as a measure of central tendency? It is less affected by outliers. 2. (2 points) Why is counterbalancing important in a within-subjects experiment? Ensuring that conditions are independent of order and of each other. This allows us to determine effect of each variable independently of the other variables. If conditions are related to order or to each other, we are unable to determine which variable is having an effect. Short answer: order effects. 3. (6 points) Define reliability and compare it to validity. Give an example of when a measure could be valid but not reliable. 2 points: Reliability is the consistency or dependability of a measurement technique. [“Getting the same result” was not accepted; it was too vague in that it did not specify the conditions (e.g., the same phenomenon) in which the same result was achieved.] 2 points: Validity is the extent to which a measurement procedure actually measures what it is intended to measure. 2 points: Example (from class) is a weight scale that gives a different result every time the same person stands on it repeatedly. Another example: a scale that actually measures hunger but has poor test-retest reliability. [Other examples were accepted.] 4. (4 points) A consumer research company wants to compare the “coverage” of two competing cell phone networks throughout Illinois. To do so fairly, they have decided that they will only compare survey data taken from customers who are all using the same cell phone model - one that is functional on both networks and has been newly released in the last 3 months.
    [Show full text]
  • The Problem of Induction
    The Problem of Induction Gilbert Harman Department of Philosophy, Princeton University Sanjeev R. Kulkarni Department of Electrical Engineering, Princeton University July 19, 2005 The Problem The problem of induction is sometimes motivated via a comparison between rules of induction and rules of deduction. Valid deductive rules are necessarily truth preserving, while inductive rules are not. So, for example, one valid deductive rule might be this: (D) From premises of the form “All F are G” and “a is F ,” the corresponding conclusion of the form “a is G” follows. The rule (D) is illustrated in the following depressing argument: (DA) All people are mortal. I am a person. So, I am mortal. The rule here is “valid” in the sense that there is no possible way in which premises satisfying the rule can be true without the corresponding conclusion also being true. A possible inductive rule might be this: (I) From premises of the form “Many many F s are known to be G,” “There are no known cases of F s that are not G,” and “a is F ,” the corresponding conclusion can be inferred of the form “a is G.” The rule (I) might be illustrated in the following “inductive argument.” (IA) Many many people are known to have been moral. There are no known cases of people who are not mortal. I am a person. So, I am mortal. 1 The rule (I) is not valid in the way that the deductive rule (D) is valid. The “premises” of the inductive inference (IA) could be true even though its “con- clusion” is not true.
    [Show full text]
  • 1 a Bayesian Analysis of Some Forms of Inductive Reasoning Evan Heit
    1 A Bayesian Analysis of Some Forms of Inductive Reasoning Evan Heit University of Warwick In Rational Models of Cognition, M. Oaksford & N. Chater (Eds.), Oxford University Press, 248- 274, 1998. Please address correspondence to: Evan Heit Department of Psychology University of Warwick Coventry CV4 7AL, United Kingdom Phone: (024) 7652 3183 Email: [email protected] 2 One of our most important cognitive goals is prediction (Anderson, 1990, 1991; Billman & Heit, 1988; Heit, 1992; Ross & Murphy, 1996), and category-level information enables a rich set of predictions. For example, you might not be able to predict much about Peter until you are told that Peter is a goldfish, in which case you could predict that he will swim and he will eat fish food. Prediction is a basic element of a wide range of everyday tasks from problem solving to social interaction to motor control. This chapter, however, will focus on a narrower range of prediction phenomena, concerning how people evaluate inductive “syllogisms” or arguments such as the following example: Goldfish thrive in sunlight --------------------------- Tunas thrive in sunlight. (The information above the line is taken as a premise which is assumed to be true, then the task is to evaluate the likelihood of the conclusion, below the line.) Despite the apparent simplicity of this task, there are a variety of interesting phenomena that are associated with inductive arguments. Taken together, these phenomena reveal a great deal about what people know about categories and their properties, and about how people use their general knowledge of the world for reasoning.
    [Show full text]
  • What Is Statistic?
    What is Statistic? OPRE 6301 In today’s world. ...we are constantly being bombarded with statistics and statistical information. For example: Customer Surveys Medical News Demographics Political Polls Economic Predictions Marketing Information Sales Forecasts Stock Market Projections Consumer Price Index Sports Statistics How can we make sense out of all this data? How do we differentiate valid from flawed claims? 1 What is Statistics?! “Statistics is a way to get information from data.” Statistics Data Information Data: Facts, especially Information: Knowledge numerical facts, collected communicated concerning together for reference or some particular fact. information. Statistics is a tool for creating an understanding from a set of numbers. Humorous Definitions: The Science of drawing a precise line between an unwar- ranted assumption and a forgone conclusion. The Science of stating precisely what you don’t know. 2 An Example: Stats Anxiety. A business school student is anxious about their statistics course, since they’ve heard the course is difficult. The professor provides last term’s final exam marks to the student. What can be discerned from this list of numbers? Statistics Data Information List of last term’s marks. New information about the statistics class. 95 89 70 E.g. Class average, 65 Proportion of class receiving A’s 78 Most frequent mark, 57 Marks distribution, etc. : 3 Key Statistical Concepts. Population — a population is the group of all items of interest to a statistics practitioner. — frequently very large; sometimes infinite. E.g. All 5 million Florida voters (per Example 12.5). Sample — A sample is a set of data drawn from the population.
    [Show full text]
  • Structured Statistical Models of Inductive Reasoning
    CORRECTED FEBRUARY 25, 2009; SEE LAST PAGE Psychological Review © 2009 American Psychological Association 2009, Vol. 116, No. 1, 20–58 0033-295X/09/$12.00 DOI: 10.1037/a0014282 Structured Statistical Models of Inductive Reasoning Charles Kemp Joshua B. Tenenbaum Carnegie Mellon University Massachusetts Institute of Technology Everyday inductive inferences are often guided by rich background knowledge. Formal models of induction should aim to incorporate this knowledge and should explain how different kinds of knowledge lead to the distinctive patterns of reasoning found in different inductive contexts. This article presents a Bayesian framework that attempts to meet both goals and describe 4 applications of the framework: a taxonomic model, a spatial model, a threshold model, and a causal model. Each model makes probabi- listic inferences about the extensions of novel properties, but the priors for the 4 models are defined over different kinds of structures that capture different relationships between the categories in a domain. The framework therefore shows how statistical inference can operate over structured background knowledge, and the authors argue that this interaction between structure and statistics is critical for explaining the power and flexibility of human reasoning. Keywords: inductive reasoning, property induction, knowledge representation, Bayesian inference Humans are adept at making inferences that take them beyond This article describes a formal approach to inductive inference the limits of their direct experience.
    [Show full text]
  • A Philosophical Treatise of Universal Induction
    Entropy 2011, 13, 1076-1136; doi:10.3390/e13061076 OPEN ACCESS entropy ISSN 1099-4300 www.mdpi.com/journal/entropy Article A Philosophical Treatise of Universal Induction Samuel Rathmanner and Marcus Hutter ? Research School of Computer Science, Australian National University, Corner of North and Daley Road, Canberra ACT 0200, Australia ? Author to whom correspondence should be addressed; E-Mail: [email protected]. Received: 20 April 2011; in revised form: 24 May 2011 / Accepted: 27 May 2011 / Published: 3 June 2011 Abstract: Understanding inductive reasoning is a problem that has engaged mankind for thousands of years. This problem is relevant to a wide range of fields and is integral to the philosophy of science. It has been tackled by many great minds ranging from philosophers to scientists to mathematicians, and more recently computer scientists. In this article we argue the case for Solomonoff Induction, a formal inductive framework which combines algorithmic information theory with the Bayesian framework. Although it achieves excellent theoretical results and is based on solid philosophical foundations, the requisite technical knowledge necessary for understanding this framework has caused it to remain largely unknown and unappreciated in the wider scientific community. The main contribution of this article is to convey Solomonoff induction and its related concepts in a generally accessible form with the aim of bridging this current technical gap. In the process we examine the major historical contributions that have led to the formulation of Solomonoff Induction as well as criticisms of Solomonoff and induction in general. In particular we examine how Solomonoff induction addresses many issues that have plagued other inductive systems, such as the black ravens paradox and the confirmation problem, and compare this approach with other recent approaches.
    [Show full text]