Tuesday, March 22, 2011


Heteroskedasticity (Value of Error Term Are Not Constant)

-Violates classical assumption 5, that errors have constant variance
(Var( i ) = σ2, ∀i )
-Not always realistic: if we put basketball players and mice into the
same dataset, most likely errors for basketball players would have
larger variance than errors for mice (if measured in the same units)
-Most likely to take place in cross-sectional models

-OLS estimates of ˆβ remain unbiased but have similar problems to
serial correlation


A classic example of heteroscedasticity is that of income versus expenditure on meals. As one's income increases, the variability of food consumption will increase. A poorer person will spend a rather constant amount by always eating less expensive food; a wealthier person may occasionally buy inexpensive food and at other times eat expensive meals. Those with higher incomes display a greater variability of food consumption.

There are several methods to test for the presence of heteroscedasticity:

1) Park  test 
2) White test
3) Breusch-Pagan test

Remedies for heteroskedasticity

 1)  Weighted least squares
 2)  Heteroskedasticity-corrected standard errors
 3)  Redefining the variables

To Detect Heteroscedasticity:
     Plot  values of residual againts value of independent  variables..

     Increase sample size data @ by  droping one of the highly correlated variables..

No comments: