Applied Regression Analysis for Business: Tools, Traps and Applications

Preface ........................................................................................................................... 1
Chapter 1 - Basics of regression models .................................................................. 2 1.1. Types and applications of regression models. .............................................................................. 2 1.2. Basic elements of a single-equation linear regression model. ..................................................... 4 Chapter 2 - Relevance of outlying and influential observations for regression analysis ..................................................................................................... 7 2.1. Nature and dangers of univariate and multivariate outlying observations. ................................ 7 2.2. Tools for detection of outlying observations. ............................................................................. 19 2.3. Recommended procedure for detection of outlying and influential observations. .................... 32 2.4. Dealing with detected outlying and influential observations. .................................................... 33 Chapter 3 - Basic procedure for multiple regression model building ............. 35 3.1. Introduction. ............................................................................................................................... 35 3.2. Preliminary specification of the model. ...................................................................................... 35 3.3. Detection of potential outliers in the dataset. ........................................................................... 40 3.4. Selection of explanatory variables (from the set of candidates). ............................................... 48 3.5. Interpretation of the obtained regression' structural para meters. ............................................ 57 Chapter 4 - Verification of multiple regression model ...................................... 60 4.1. Introduction. ............................................................................................................................... 60 4.2. Testing general statistical significance of the whole model: F test. ........................................... 61 4.3. Testing the normality of regression residuals' distribution. ....................................................... 63 4.4. Testing the autocorrelation of regression residuals. .................................................................. 72 4.5. Testing the heteroscedasticity of regression residuals. .............................................................. 87 4.6. Testing the symmetry of regression residuals. ....................................................................... .... 97 4.7. Testing the randomness of regression residuals. ..................................................................... 106 4.8. Testing the specification of the model: Ramsey's RESET test. ................................................. 115 4.9. Testing the multicollinearity of explanatory variables. ............................................................ 121 4.10. What to do if the model is not correct? .................................................................................. 125 4.11. Summary of verification of our model .................................................................................... 130 Chapter 5 - Common adjustments to multiple regressions .............................. 132 5.1. Dealing with qualitative factors by means of dummy variables. ............................................. 132 5.2. Modeling seasonality by means of dummy variables. ....... ...................................................... 136 5.3. Using dummy variables for outlying observations. .................................................................. 148 281 5.4. Dealing with structural changes in modeled relationships. ..................................................... 155