Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: R: Goodness of fit measure akin to R-squared for 0-constant or noconstant


From   Bas de Goei <bas.degoei@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: R: Goodness of fit measure akin to R-squared for 0-constant or noconstant
Date   Fri, 24 Apr 2009 11:57:43 +0100

Cheers from the UK!

On Fri, Apr 24, 2009 at 11:25 AM, Martin Weiss <martin.weiss1@gmx.de> wrote:
> <>
>
>
> I do think that such a -relatively- easy formula could be calculated in the
> old -matrix- commands (see -help matrix-). All it takes would be transposing
> the (Y-XBhat) matrix, subtraction and multiplication. The -help matrix- for
> Stata 10 contains an example for the OLS estimator calculation, that reads
> as
>
> ***
> matrix beta = invsym(X'*X)*X'*y
> ***
>
> so that is pretty much what you need...
>
> HTH
> Martin
>
>
> -----Ursprüngliche Nachricht-----
> Von: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Bas de Goei
> Gesendet: Freitag, 24. April 2009 12:20
> An: statalist@hsphsun2.harvard.edu
> Betreff: Re: st: R: Goodness of fit measure akin to R-squared for 0-constant
> or noconstant
>
> Ah...yes you're right. Y would be a (n x 1) vector...hmm, I only have
> Stata 8, any idea how y'd do such a calc without mata?
>
> On Fri, Apr 24, 2009 at 11:13 AM, Martin Weiss <martin.weiss1@gmx.de> wrote:
>> <>
>>
>> This formula screams "MATA"!
>>
>> See
> http://www.stata.com/meeting/fnasug08/baum_StataMata.beamer.FNASUG08.pdf
>>
>>
>>
>> HTH
>> Martin
>>
>>
>> -----Ursprüngliche Nachricht-----
>> Von: owner-statalist@hsphsun2.harvard.edu
>> [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Bas de Goei
>> Gesendet: Freitag, 24. April 2009 12:05
>> An: statalist@hsphsun2.harvard.edu
>> Betreff: Re: st: R: Goodness of fit measure akin to R-squared for
> 0-constant
>> or noconstant
>>
>> Hmm, my searches online have provided me with some insightful work by
>> Kvalseth (1985). Apparently, he has an alternative R-squared which
>> should work across models (including no constant or 0 constant
>> models).
>>
>> It's specified by 1 - [(Y-XBhat) ' (Y-XBhat) / Y'Y - Ymean squared]
>>
>> I could put it in myself, or is there already a user-written command
>> for this uniform R-squared?
>>
>>
>>
>> On Fri, Apr 24, 2009 at 10:50 AM, Carlo Lazzaro
>> <carlo.lazzaro@tiscalinet.it> wrote:
>>> Dear Bas,
>>> I don't know whether or not your models (with and without constant) can
> be
>>> fruitfully compared via AIC or BIC criteria.
>>>
>>> However, my knee-jerk advice is typing:
>>>
>>> - search postestimation timeseries -
>>>
>>> from within Stata.
>>>
>>> Sorry I cannot be more helpful.
>>>
>>> Kind Regards,
>>> Carlo
>>> -----Messaggio originale-----
>>> Da: owner-statalist@hsphsun2.harvard.edu
>>> [mailto:owner-statalist@hsphsun2.harvard.edu] Per conto di Bas de Goei
>>> Inviato: venerdì 24 aprile 2009 10.54
>>> A: statalist@hsphsun2.harvard.edu
>>> Oggetto: st: Goodness of fit measure akin to R-squared for 0-constant or
>>> noconstant
>>>
>>> Dear all,
>>>
>>> I am currently creating forecasts for jewellery demand in India by
>>> regressing GDP on demand for jewellery.
>>>
>>> Let me first give the required background:
>>> I have data going back to 1980. In a regression based on GDP over
>>> time, you obviously run into the problem of serial autocorrelation,
>>> though this is neccesarily a problem for a forecast, my boss wants
>>> "only regressions that pass Durbin Watson test".
>>>
>>> I really have two problems:
>>>
>>> The first is that the normal OLS regression result indicated a
>>> positive intercept. However, economically this would mean that even
>>> when there is no growth in GDP, there would still be growth in the
>>> demand for jewellery. Of course, there was the problem that the model
>>> did not pass the Durbin Watson test. Fitting the model with the GLS
>>> approach (the prais command in Stata), did improve the model, but it
>>> kept (as expected) the intercept positive.
>>>
>>> I decided to inspect the data more closely, and to drop two outliers
>>> from the data. The intercept under the Prais command is now still
>>> positive, but it has become insignificant. I decided that there is
>>> justification to re-run the regression with a 0 intercept. However,
>>> this balloons the F statistic and the R-squared. I now understand why
>>> that is, given the mathematics behind the R squared calculation.
>>>
>>> My question is, how would you calculate in Stata a "correct" or
>>> "alternative" R-squared, or a goodness of fit measure, which you can
>>> use to compare it to the model with a constant??
>>>
>>> Thanks!!
>>>
>>> Bastiaan
>>> *
>>> *   For searches and help try:
>>> *   http://www.stata.com/help.cgi?search
>>> *   http://www.stata.com/support/statalist/faq
>>> *   http://www.ats.ucla.edu/stat/stata/
>>>
>>>
>>>
>>> *
>>> *   For searches and help try:
>>> *   http://www.stata.com/help.cgi?search
>>> *   http://www.stata.com/support/statalist/faq
>>> *   http://www.ats.ucla.edu/stat/stata/
>>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>>
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index