Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

Re: st: RE: dfuller: why do I get different results?


From   Yuval Arbel <[email protected]>
To   [email protected]
Subject   Re: st: RE: dfuller: why do I get different results?
Date   Fri, 18 Nov 2011 18:31:23 +0200

Nick,

Note that I looked very carefully at the material Austin Nicols sent
me. His implication was that I should run a unit root test on a panel
dataset. I tried to run the following commands, but they did not seem
to work:

. xtset appt time_index
       panel variable:  appt (unbalanced)
        time variable:  time_index, 0 to 114, but with gaps
                delta:  1 unit

. xtunitroot llc reduct_per
Levin-Lin-Chiu test requires strongly balanced data
r(498);

. xtunitroot ht reduct_per
Harris-Tzavalis test requires strongly balanced data
r(498);

. xtunitroot breitung reduct_per
Breitung test requires strongly balanced data
r(498);




On Fri, Nov 18, 2011 at 3:08 PM, Nick Cox <[email protected]> wrote:
> See my earlier answer on using -foreach- with an FAQ as reference. I don't know if it will work, in the sense of doing what you want.
>
> By the way, I have a horrible feeling that you are in econometric peril here in some sense, and I am not endorsing your choices.
>
> You've already chosen to ignore the advice of Austin Nichols. If I were doing econometrics, I would want a very good reason to ignore Austin Nichols.
>
> By the way, Austin Nichols is not another "Nick".
>
> Nick
> [email protected]
>
> Yuval Arbel
>
> Your answer brings me to my second question:
>
> I am trying to run:
>
> bysort appt: dfuller reduct_per,noconstant regress
>
> but I'm getting the message:
>
> dfuller may not be combined with by
>
> I also tried to use -statsby- but it doesn't work either:
>
> Is there another way to run dfuller by apartments?
>
> On Fri, Nov 18, 2011 at 1:37 PM, Nick Cox <[email protected]> wrote:
>
>> bysort appt: gen reduct1=reduct_per[_n-1]
>>
>> and
>>
>> gen reduct1 = L1.reduct
>>
>> give identical results only under certain conditions. One is that sorting by -appt- does _not_ itself guarantee that values for each -appt- are sorted in time order. There can be other problems with omitted observations, etc.
>>
>> Use time-series operators after -tsset- to generate lagged variables.
>>
>> Nick
>> [email protected]
>>
>>
>> -----Original Message-----
>> From: [email protected] [mailto:[email protected]] On Behalf Of Yuval Arbel
>> Sent: 18 November 2011 11:30
>> To: statalist
>> Subject: st: dfuller: why do I get different results?
>>
>> Dear Statalist Participants,
>>
>> when I run:
>>
>> . dfuller reduct_per if appt==2862,noconstant regress
>>
>> I get the following outcome:
>>
>> Dickey-Fuller test for unit root                   Number of obs   =        37
>>
>>                               ---------- Interpolated Dickey-Fuller ---------
>>                  Test         1% Critical       5% Critical      10% Critical
>>               Statistic           Value             Value             Value
>> ------------------------------------------------------------------------------
>>  Z(t)             -6.026            -2.641            -1.950            -1.605
>>
>> ------------------------------------------------------------------------------
>> D.reduct_per |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
>> -------------+----------------------------------------------------------------
>>  reduct_per |
>>         L1. |  -.5409015   .0897625    -6.03   0.000    -.7229484   -.3588546
>> ------------------------------------------------------------------------------
>>
>> Those outcomes imply that the calculated statistic for the unit-root
>> test is -6.03
>>
>> But when I define:
>>
>> bysort appt: gen reduct1=reduct_per[_n-1]
>> bysort appt: gen dreduct1=reduct_per-reduct_per[_n-1]
>>
>> and I run:
>>
>>
>> regress dreduct1 reduct1 if appt==2862,noconst
>>
>> I get:
>>
>> . regress dreduct1 reduct1  if appt==2862,noconst
>>
>>      Source |       SS       df       MS              Number of obs =      36
>> -------------+------------------------------           F(  1,    35) =    0.00
>>       Model |           0     1           0           Prob > F      =  1.0000
>>    Residual |         625    35  17.8571429           R-squared     =  0.0000
>> -------------+------------------------------           Adj R-squared = -0.0286
>>       Total |         625    36  17.3611111           Root MSE      =  4.2258
>>
>> ------------------------------------------------------------------------------
>>    dreduct1 |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
>> -------------+----------------------------------------------------------------
>>     reduct1 |          0   .0509647     0.00   1.000    -.1034639    .1034639
>> ------------------------------------------------------------------------------
>>
>> .
>> Shouldn't I get exactly the same outcomes in both regressions?
>
> *
> *   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/
>



-- 
Dr. Yuval Arbel
School of Business
Carmel Academic Center
4 Shaar Palmer Street, Haifa, Israel
e-mail: [email protected]

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index