Econometrics and Data Analysis for Developing Countries

by Mukherjee; Chandan

• ISBN13: 9780415093996
• ISBN10: 0415093996
• Edition: Disk
• Format: Hardcover
• Copyright: 1998-01-16
• Publisher: Routledge

Summary

Getting accurate data on less developed countries has created great problems for studying these areas. Yet until recently students of development economics have relied on standard econometrics texts, which assume a Western context.Econometrics and Data Analysis forDeveloping Countriessolves this problem. It will be essential reading for all advanced students of development economics.

Table of Contents

List of figures	ix	(4)
List of tables	xiii	(3)
List of boxes	xvi	(1)
Preface	xvii
Introduction	1	(20)
1 The purpose of this book	1	(2)
2 The approach of this book: an example	3	(18)
Part I Foundations of data analysis	21	(88)
1 Model specification and applied research	23	(21)
1.1 Introduction	23	(1)
1.2 Model specification and statistical inference	24	(5)
1.3 The role of data in model specification: traditional modelling	29	(3)
1.4 The role of data in model specification: modern approaches	32	(7)
1.5 The time dimension in data	39	(3)
1.6 Summary of main points	42	(2)
2 Modelling an average	44	(31)
2.1 Introduction	44	(1)
2.2 Kinds of averages	45	(5)
2.3 The assumptions of the model	50	(3)
2.4 The sample mean as best linear unbiased estimator (BLUE)	53	(5)
2.5 Normality and the maximum likelihood principle	58	(3)
2.6 Inference from a sample of a normal distribution	61	(10)
2.7 Summary of main points	71	(2)
Appendix 2.1: Properties of mean and variance	73	(1)
Appendix 2.2: Standard sampling distributions	73	(2)
3 Outliers, skewness and data transformations	75	(34)
3.1 Introduction	75	(1)
3.2 The least squares principle and the concept of resistance	76	(4)
3.3 Mean-based versus order-based sample statistics	80	(10)
3.4 Detecting non-normality in data	90	(7)
3.5 Data transformations to eliminate skewness	97	(9)
3.6 Summary of main points	106	(3)
Part II Regression and data analysis	109	(140)
4 Data analysis and simple regression	111	(52)
4.1 Introduction	111	(1)
4.2 Modelling simple regression	112	(2)
4.3 Linear regression and the least squares principle	114	(6)
4.4 Inference from classical normal linear regression model	120	(4)
4.5 Regression with graphics: checking the model assumptions	124	(12)
4.6 Regression through the origin	136	(1)
4.7 Outliers, leverage and influence	137	(11)
4.8 Transformation towards linearity	148	(11)
4.9 Summary of main points	159	(4)
5 Partial regression: interpreting multiple regression coefficients	163	(45)
5.1 Introduction	163	(2)
5.2 The price of food and the demand for manufactured goods in India	165	(8)
5.3 Least squares and the sample multiple regression line	173	(7)
5.4 Partial regression and partial correlation	180	(4)
5.5 The linear regression model	184	(8)
5.6 The t-test in multiple regression	192	(6)
5.7 Fragility analysis: making sense of regression coefficients	198	(8)
5.8 Summary of main points	206	(2)
6 Model selection and misspecification in multiple regression	208	(41)
6.1 Introduction	208	(1)
6.2 Griffin's aid versus savings model: the omitted variable bias	209	(3)
6.3 Omitted variable bias: the theory	212	(7)
6.4 Testing zero restrictions	219	(10)
6.5 Testing non-zero linear restrictions	229	(2)
6.6 Tests of parameter stability	231	(6)
6.7 The use of dummy variables	237	(9)
6.8 Summary of main points	246	(3)
Part III Analysing cross-section data	249	(84)
7 Dealing with heteroscedasticity	251	(28)
7.1 Introduction	251	(1)
7.2 Diagnostic plots: looking for heteroscedasticity	252	(4)
7.3 Testing for heteroscedasticity	256	(8)
7.4 Transformations towards homoscedasticity	264	(6)
7.5 Dealing with genuine heteroscedasticity: weighted least squares and heteroscedastic standard errors	270	(7)
7.6 Summary of main points	277	(2)
8 Categories, counts and measurements	279	(23)
8.1 Introduction	279	(1)
8.2 Regression on a categorical variable: using dummy variables	280	(7)
8.3 Contingency tables: association between categorical variables	287	(6)
8.4 Partial association and interaction	293	(2)
8.5 Multiple regression on categorical variables	295	(3)
8.6 Summary of main points	298	(4)
9 Logit transformation, modelling and regression	302	(31)
9.1 Introduction	302	(1)
9.2 The logit transformation	303	(4)
9.3 Logit modelling with contingency tables	307	(6)
9.4 The linear probability model versus logit regression	313	(7)
9.5 Estimation and hypothesis testing in logit regression	320	(7)
9.6 Graphics and residual analysis in logit regression	327	(4)
9.7 Summary of main points	331	(2)
Part IV Regression with time-series data	333	(80)
10 Trends, spurious regressions and transformations to stationarity	335	(31)
10.1 Introduction	335	(1)
10.2 Stationarity and non-stationarity	335	(3)
10.3 Random walks and spurious regression	338	(11)
10.4 Testing for stationarity	349	(7)
10.5 Transformations to stationarity	356	(7)
10.6 Summary of main points	363	(2)
Appendix 10.1: Generated DSP and TSP series for exercises	365	(1)
11 Misspecification and autocorrelation	366	(27)
11.1 Introduction	366	(1)
11.2 What is autocorrelation and why is it a problem?	366	(4)
11.3 Why do we get autocorrelation?	370	(9)
11.4 Detecting autocorrelation	379	(8)
11.5 What to do about autocorrelation	387	(3)
11.6 Summary of main points	390	(1)
Appendix 11.1: Derivation of variance and covariance for AR(1) model	391	(2)
12 Cointegration and the error correction model	393	(20)
12.1 Introduction	393	(1)
12.2 What is cointegration?	393	(6)
12.3 Testing for cointegration	399	(7)
12.4 The error correction model (ECM)	406	(6)
12.5 Summary of main points	412	(1)
Part V Simultaneous equation models	413	(42)
13 Misspecification bias from single equation estimation	415	(22)
13.1 Introduction	415	(2)
13.2 Simultaneity bias in a supply and demand model	417	(5)
13.3 Simultaneity bias: the theory	422	(3)
13.4 The Granger and Sims tests for causality and concepts of exogeneity	425	(3)
13.5 The identification problem	428	(6)
13.6 Summary of main points	434	(3)
14 Estimating simultaneous equation models	437	(18)
14.1 Introduction	437	(1)
14.2 Recursive models	437	(2)
14.3 Indirect least squares	439	(3)
14.4 Instrumental variable estimation and two-stage least squares	442	(3)
14.5 Estimating the consumption function in a simultaneous system	445	(3)
14.6 Full information estimation techniques	448	(3)
14.7 Summary of main points	451	(4)
Appendix A: The data sets used in this book	455	(8)
Appendix B: Statistical tables	463	(18)
References	481	(4)
Index	485