ABSTRACT
In Chapter 6 we present an overview of regression techniques. We have found in our advising that students (future decision-makers) often use the wrong technique in the work. Here, we present simple linear regression, multiple linear regression, nonlinear regression, logistics regression, and Poisson regression. The approach here is when to use each type of regression based on the data available. A main real example is taken from the work done for the government of the Philippines; it has collected data on violence acts committed by terrorists and insurgents over the past decade. Over the same period they have collected data on the population such as education levels (literacy), employment, government satisfaction, ethnicity, and such. The government is looking to see what it can do as improvements might reduce the number of violent acts committed. What should the government do to improve the situation? In this chapter, we will briefly discuss some regression techniques as background information and point keys to finding adequate models. We do not try to cover all the regression topics that exist but we do illustrate some real examples and the techniques used to gain insights, predict, explain, and answer scenario-related questions. We confine ourselves to simple linear regression, multiple regression, nonlinear regression (exponential and sine), binary logistics regression, and Poisson regression.
