Assumptions for Multiple linear regression

  1. Relationship between dependent variable and independent variables is linear.
  2. There is no linear relationship between independent variables and they are also not random. If there is linear relationship between independent variables, it is called Multicollinearity leading to high R^2 and significant F Stat. This results in inflated Std Err and low T Stat ( opposite of heteroskedasticity effect).
  3. Expected value of Error term is 0 and Error term is normally distributed.
  4. Variance of Error term is the same for all observations. If this assumption is not true, it is called Heteroskedasticity.This can be tested by Breuch Pagan test. If there is Heteroskedasticity . , then Fstat is unreliable and SError is understated and T stat is overstated.
  5. Error term is uncorrelated across observations. If it is correlated, then it is called Serial Correlation. This is tested using Durban Watson (DW) test. Here F stat and T stat is too high. DW = 2(1-r)
Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s