Narrow Results

The following sections are included:

• Introduction

• Resource

• Generalized Sedyakin's model

• Additive accumulation of damages model

• Proportional hazards model

• Generalized proportional hazards models

• The main classes of GPH models.

• Parametrization of the function r in AAD and GPH models

• Changing shape and scale models

• Generalizations

• The heredity hypothesis

• References

The following sections are included:

• INTRODUCTION

• MULTICOLLINEARITY
  • Definition and Effect
  • Rules of Thumb in Determining the Degree of Collinearity

• HETEROSCEDASTIC1TY
  • Definition and Concept
  • Evaluating the Existence of Heteroscedasticity

• AUTOCORRELATION
  • Basic Concept
  • The Durbin–Watson Statistic

• MODEL SPECIFICATION AND SPECIFICATION BIAS (Optional)
  • Impact of the Omission of Variables on Estimated Regression Coefficients

• NONLINEAR MODELS (Optional)
  • The Quadratic Model
  • The Log-Linear and the Log-Log Linear Model

• LAGGED DEPENDENT VARIABLES (Optional)
  • The Relationship Between Dividend per Share and Earnings per Share
  • Consumption Function Analysis

• DUMMY VARIABLES
  • Analysis of the Money Supply

• REGRESSION WITH INTERACTION VARIABLES
  • Analysis of the Money Supply, with Interaction

• REGRESSION APPROACH TO INVESTIGATING THE EFFECT OF ALTERNATIVE BUSINESS STRATEGIES⁶

• Summary

• Appendix 16A Dynamic Ratio Analysis

• Appendix 16B Term Structure of Interest Rate

• Questions and Problems

• Project IV Project for Regression and Correlation Analyses

The purpose of this article is to introduce an example of statistical estimation based on conditional likelihood for stratified educational data. The example is based on the data that was presented in Ref. 2. The conditional likelihood estimation on stratified categorical data requires marginally constrained data enumeration, which is closely related to identifying Markov basis.