Regression methods
As we learned, regression allows us to model the relationship between two or more variables, especially when a continuous dependent variable is predicted, based on several independent variables. The independent variables used in regression can be either continuous or dichotomous. In cases where the dependent variable is dichotomous, logistic regression is applied. In cases where the split between the two levels of dependent variables is equal, then both linear and logistic regression would fetch the same results.
Assumptions of regression (most apply to linear regression model family)
- Sample cases size: In order to apply regression models, the cases-to-Independent Variables (IVs) ratio should ideally be 20:1 (for every IV in the model, there need to be 20 cases), the least being 5:1(5 cases for every IV in the model).
- Data accuracy: Regression assumes the basic validity of data, and it is expected to run basic data validations before running regression methods. For example...
