The Measurement of Economic Relationships

Astranger in academia cannot but be impressed by the apparent uniformity and precision of the methodology currently applied to the measurement of economic relationships. In scores of journal articles and other studies, a theoretical argument is typically presented to justify the position that a certain variable is related to certain other, possibly causal, variables. Regression or a related method is applied to a set of observations on these variables, and the conclusion often emerges that the causa,l variables are indeed "significant" at a certain "level," thereby lending support to the theoretical argument-an argument presumably formulated independently of the observations. A variable may be declared significant (and few doubt that this does not mean important) at, say, the 0. 05 level, but not the 0. 01. The effects of the variables are calculated to many significant digits, and are often accompanied by intervals and forecasts of not quite obvious meaning but certainly of reassuring "confidence. " The uniformity is also evident in the many mathematically advanced text­ books of statistics and econometrics, and in their less rigorous introductory versions for students in economics or business. It is reflected in the tools of the profession: computer programs, from the generaiones addressed to the incidental researcher to the dedicated and sophisticated programs used by the experts, display the same terms and implement the same methodology. In short, there appears no visible alternative to the established methodol­ ogy and no sign of reservat ions concerning its validity.

Contenu

Preface Chapter 1 Introduction 1.1 The Status Quo 1.2 The CLM in Academic Studies 1.3 The CLM in Practice 1.4 Extensions of the CLM 1.5 The Road Ahead Chapter 2 The Fitting Method: An Introduction 2.1 Introduction 2.2 The Problem 2.3 The Available Information 2.4 One Solution 2.5 Least Squares and Spreadsheets 2.6 Constrained Least Squares 2.7 Tolerance Intervals 2.8 Joint Tests and Tolerance Regions 2.9 Interval Forecasts 2.10 Computer Output 2.11 In Summary Chapter 3 The Fitting Method: A formal Treatment 3.1 Introduction 3.2 Relationships 3.3 Unrestricted Least Squares 3.4 Restricted Least Squares 3.5 Ordinary Tolerance Intervals and Regions 3.6 A Tolerance Region for All Parameters 3.7 Tolerance Interval Forecasts 3.8 Possible Extensions Chapter 4 The Classical Linear Model 4.1 Introduction 4.2 The Assumptions of the CLM 4.3 Estimates and Their Properties 4.4 Statistical Inference 4.5 Specification Error 4.6 On Confidence Interval Estimates 4.7 The Many Problems of Significance 4.8 On Confidence Interval Forecasts 4.9 The Art and Practice of Statistical Inference 4.10 Bad Practice or Bad Theory? Chapter 5 The Central Assumptions 5.1 Introduction 5.2 True Parameters? 5.3 The Randomness of Error 5.4 Probability 5.5 The CentralLimit Theorem and Normality 5.6 Are the Unknown Factors Random Variables? 5.7 Serial Correlation 5.8 The 'As If' Argument 5.9 A Probably Deviation 5.10 On the Distribution of Residuals 5.11 In Summary Chapter 6 Random Processes 6.1 Introduction 6.2 The Coin Toss 6.3 Of Birth and Deaths 6.4 Stock market Prices 6.5 Some Perils of Time Series Analysis 6.6 In Conclusion Chapter 7 The 'Probabilistic Revolution' 7.1 Introduction 7.2 Before Haavelmo 7.3 Haavelmo on Relationships 7.4 Haavelmo in Contemporary Reviews 7.5 The Probability Approach Reconsidered 7.6 Random Sampling 7.7 The Assumptions Reconsidered, Continuation 7.8 In Summary Chapter 8 Assessment 8.1 The Fitting Method in Perspective 8.2 The Tolerance Level 8.3 The Technical Pursuit of Fit 8.4 The Success Rate of Tolerance Interval Forecasts 8.5 The Poverty of Properties 8.6 Does it Matter 8.7 Subjective Probability 8.8 Determinism and Probabilism 8.9 The 'As If' Assumption Revisited 8.10 Why the Status Quo 8.11 A Pragmatic Approach References Index