|From||Goodall S <S.Goodall@rhul.ac.uk>|
|Subject||st: Date: Tue, 15 Mar 2005 18:52:24 -0000|
|Date||Tue, 15 Mar 2005 13:52:34 -0500 (EST)|
Dear members of Statlist,
I am very new to this system and similarly new to Stata (at least at an advanced level). I am attempting to develop a model for my dissertation.I would like to develop a very simple model with house prices as my left-hand side variable and rent, income, mortgage rent, and CPI as my right hand-side variables. I have collected my data, formatted my data (including putting several variables in real terms) and found that I have the best model available to me (in terms of ommited variables at least). I have come to the conclusion that my model has no ommitted variables and is normal, but does suffer from autocorrelation, heteroscedasticity, and multicollinearity. Is it true that I can correct for autocorrelation and heteroscedasticity using a Newey West test? If so, would you be so kind as to give me specific details of how to perform this test and how to interpret its results? Also, any other suggestions or help would be much appreciated. I know this must seem like play stuff to you guys but I really have been stuck with certain aspects of Stata for some time now. I also appreciate that I probably cannot be of much use to any of you when it comes to Stata. Henceforth, if there is anything else I can do to help any one of you please do not hesitate to ask.
Many thanks, Scott