Assumptions of ordinary least-squares estimation
The regression model can be used to describe the relationships between two or more variables in a sample without making any assumptions except that the dependent variable is continuous and that the relationships between these variables are linear. As a result, regression models can be used almost anytime in a purely descriptive manner to summarize the relationships between the variables in a sample. However, it is not possible to make valid statistical inferences about population parameters from sample statistics without making at least some assumptions. After all, statisticians must make certain assumptions about the characteristics of a population in order to derive the sampling distributions of the sample statistics drawn from that population. Fortunately, most of the assumptions associated with regression analysis are relatively weak in the sense that they are quite reasonable in most cases.
KeywordsRegression Coefficient Sample Statistic Central Limit Theorem Sampling Distribution Sample Estimate
Unable to display preview. Download preview PDF.