Heteroscedasticity

[Guest]

Hi!
I am having some issues with a linear regression analysis. I have heteroscedasticity in my model and don't know how to fix it. Did the esat hettest wich came out significant. This is the image of the two way scatterplot.

Scatter m3.gph

I am not very good at diagnosing hetero and how serious it is. Is this bad enough that I cannot trust the results in the analysis?

[Carlo Lazzaro]

Guest:
as per your graph, the heteroskedasticity, if any, deos not look severe.
You may want to run -regress- with the robust option and compare the standadrd errors with the default ones.