Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Apologies for request

From   Mark Schaffer <[email protected]>
To   [email protected], Goodall S <[email protected]>
Subject   Re: st: Apologies for request
Date   Tue, 15 Mar 2005 23:11:41 +0000 (GMT)


I think you might have misunderstood Nick's point.  When he says that most
of your questions require recourse to a standard text, he probably means,
not a Stata text or the manuals, but a standard econometrics or statistics
textbook (e.g., on how to calculate Newey-West-style standard errors - their
procedure is not a test).

BTW, if you believe you have autocorrelation, heteroskedasticity, and
multicollinearity problems, it's possible that your omitted variable test
isn't valid (isn't robust in the presence of the problems).


Quoting Goodall S <[email protected]>:

> Unfortunately, although I have spent just about every waking minute
> for the
> past week in front of a screen with Stata 8.2 imprinted on it I
> haven't been
> able to grasp its capabilities very well. I have not found
> information on
> the net or in the textbooks conclusive or straghtforward.
> Furthermore, I do
> not have a copy of the Stata manual nor have time enough to wait for
> it to
> arrive as my work is to finished very soon. I'm sorry to have
> bothered you
> all for some simple advice, I did not realize it would cause such a
> problem
> Scott
> Nick Cox
> Please don't send formatted text to Statalist
> Please do use sensible titles. These points,
> and other well-meant advice, are explained
> in the Statalist FAQ, the URL of which is
> at the bottom of every Statalist posting.
> There is a Newey-West command in Stata: its
> help is accessible at -whelp newey-
> and there is more in the manuals at [TS] newey.
> Most of your other questions seem to require
> recourse to a standard text, at least in
> the first instance.
> Nick
> [email protected]
> Scott Goodall
> 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.
> *
> *   For searches and help try:
> *
> *
> *
> *
> *   For searches and help try:
> *
> *
> *

Prof. Mark Schaffer
Director, CERT
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS
tel +44-131-451-3494 / fax +44-131-451-3294
email: [email protected]


This message is subject to 


*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index