Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: r(198) error using lrtest to test for heteroskedasticity.


From   Sean Lim <seanlim174@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: r(198) error using lrtest to test for heteroskedasticity.
Date   Sat, 7 Jul 2012 03:42:28 +0100

That explains it all.  Thank you for making this so much clearer to me!

On 7 July 2012 01:44, Nick Cox <njcoxstata@gmail.com> wrote:
> You omitted to estimate the model again. Therefore -hetero- is still
> the most recent set of estimates and you are comparing it with itself.
> The -lrtest- hinges on comparing two different model fits. If you look
> again at the FAQ you will see that there are two model fits.
>
> (I first looked at your bottom line, which looked wrong. I didn't
> notice in my first posting that you made two mistakes.)
>
> Nick
>
> On Sat, Jul 7, 2012 at 12:07 AM, Sean Lim <seanlim174@gmail.com> wrote:
>> Dear Nick,
>>
>> Thank you for your reply.
>>
>> I have tried the commands with the alterations you suggested.
>>
>> I am afraid I still have the same error even when I include the period(.).
>>
>> . lrtest hetero . , df(20)
>> models hetero specified more than once
>> r(198);
>>
>> Could this be an issue with my dataset? I have a very unbalanced panel
>> with many data points missing in the independent variables.
>>
>> On 6 July 2012 19:24, Nick Cox <njcoxstata@gmail.com> wrote:
>>> The FAQ recommends different syntax. You omitted the period (.) in
>>>
>>> lrtest hetero . , df(`df')
>>>
>>> Nick
>>>
>>> On Fri, Jul 6, 2012 at 1:15 PM, Sean Lim <seanlim174@gmail.com> wrote:
>>>> Dear Statalisters,
>>>>
>>>> I am trying to test for heteroskedasticity in a panel dataset I have.
>>>> I am following the steps described in
>>>> http://www.stata.com/support/faqs/statistics/panel-level-heteroskedasticity-and-autocorrelation/
>>>>  but I am getting an error: " r(198) error when trying to use lrtest.
>>>>
>>>>
>>>> Any advice on how I should continue with this is appreciated. Below is
>>>> my output.
>>>>
>>>> Many thanks,
>>>>
>>>>
>>>> Sean
>>>>
>>>> . xtgls leverage avol7 time_tomaturity grade1 bid_ask_predict, igls
>>>> panels(heteroskedastic)
>>>> Iteration 1: tolerance = .03519474
>>>> Iteration 2: tolerance = .10619771
>>>> Iteration 3: tolerance = .13432693
>>>> Iteration 4: tolerance = .12540942
>>>> Iteration 5: tolerance = .0969322
>>>> Iteration 6: tolerance = .02952538
>>>> Iteration 7: tolerance = .01171261
>>>> Iteration 8: tolerance = .00374088
>>>> Iteration 9: tolerance = .00121751
>>>> Iteration 10: tolerance = .00043942
>>>> Iteration 11: tolerance = .00016508
>>>> Iteration 12: tolerance = .00006333
>>>> Iteration 13: tolerance = .00002464
>>>> Iteration 14: tolerance = 9.688e-06
>>>> Iteration 15: tolerance = 3.844e-06
>>>> Iteration 16: tolerance = 1.537e-06
>>>> Iteration 17: tolerance = 6.189e-07
>>>> Iteration 18: tolerance = 2.506e-07
>>>> Iteration 19: tolerance = 1.019e-07
>>>> Iteration 20: tolerance = 4.167e-08
>>>>
>>>>
>>>> Cross-sectional time-series FGLS regression
>>>>
>>>> Coefficients:  generalized least squares
>>>> Panels:        heteroskedastic
>>>> Correlation:   no autocorrelation
>>>>
>>>> Estimated covariances      =        21          Number of obs      =     15018
>>>> Estimated autocorrelations =         0          Number of groups   =        21
>>>> Estimated coefficients     =         5          Obs per group: min =       110
>>>>                                                                avg =  715.1429
>>>>                                                                max =       782
>>>>                                                 Wald chi2(4)       = 953523.91
>>>> Log likelihood             =  13896.44          Prob > chi2        =    0.0000
>>>>
>>>> ------------------------------------------------------------------------------
>>>>     leverage |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
>>>> -------------+----------------------------------------------------------------
>>>>        avol7 |  -.7824862   .0049243  -158.90   0.000    -.7921376   -.7728347
>>>> time_tomat~y |  -.0002712   .0000154   -17.65   0.000    -.0003013   -.0002411
>>>>       grade1 |   -.399934   .0010845  -368.76   0.000    -.4020597   -.3978084
>>>> bid_ask_p~ct |   .0661513   .0043105    15.35   0.000     .0577029    .0745996
>>>>        _cons |   .9815676   .0009443  1039.42   0.000     .9797167    .9834184
>>>> ------------------------------------------------------------------------------
>>>>
>>>> . estimates store hetero
>>>>
>>>> . local df = e(N_g) - 1
>>>>
>>>> . di `df'
>>>> 20
>>>>
>>>> . lrtest hetero , df(20)
>>>> models hetero specified more than once
>>>> r(198);
>>> *
>>> *   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   |   Site index