Some guidelines for regression analysis

# Some guidelines for regression analysis - Some guidelines...

This preview shows pages 1–2. Sign up to view the full content.

Some guidelines for regression analysis (1) Get descriptive statistics of all variables, including mean, std, dev., min, max and a histogram. Get a “feeling” of the data first. (2)Run scatter plots of all pairs of variables. The dependent variable y should be on the y-axis. Get a preliminary feel of what functional form the response variable y is related to the independent variables. (3)Get a correlation matrix, check to see if some independent variables are highly correlated, and decide whether there’s multicollinearity in the data. (4)Run a first order model and get the residual plot and histogram of the residuals. (5)Run a second order model and repeat the same as (4). The following are general procedures for regression analysis with less than 10 variables. When there are more than 10 variables, we need to perform step wise selection to determine the most important variables. (1) We look at the results of the first and second order models.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 09/28/2011 for the course STAT METHO 33:623:385 taught by Professor Faridalizadeh during the Spring '11 term at Rutgers.

### Page1 / 3

Some guidelines for regression analysis - Some guidelines...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online