Now we will deal with some more specialized topics that are usually not covered in an introductory course series. Some of these topics require more specialized mathematical knowledge than the analysis of multiple regression in Parts 1 and 2. In Chapter 13, we will show how to apply multiple regression to independently grouped stratifications. The topics discussed are very similar to classical stratified analysis, with the difference that we can study how relationships change over time by incorporating time dummy variables. We will also show how stratified time series data (panel data) can be analyzed within a regression framework. Chapter 14 examines more advanced methods for panel data that are nonetheless systematically used in applications. Chapters 15 and 16 explore the problem of endogenous explanatory variables. In Chapter 15, we introduce the method of instrumental variables as a method for solving the problem of omitted variables, as well as the problem of measurement error. The two-stage least squares method is quite often used in empirical economics and is essential for estimating systems of equations models, a topic we will examine in Chapter 16. Chapter 17 examines some quite advanced topics that are usually used in stratified analysis, including models for limited dependent variables and methods for correcting sample selection bias. Chapter 18 turns in a different direction by examining some recent developments in time series econometrics that have proven useful for estimating dynamic relationships. Chapter 19 can help students who need to write a paper in the field of social sciences. Suggestions are made on how to choose a topic, collect and analyze data, and write the paper.