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: what does increase in coefficient value over two time period indicate


From   Maarten Buis <maartenlbuis@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: what does increase in coefficient value over two time period indicate
Date   Mon, 5 Nov 2012 13:28:15 +0100

I obviously cannot tell you what you should consider "big" or
"important"; only you can do that. The statistical test is also
something only you can do, as you have the data. So there is not much
more I can say other than "have fun".

-- Maarten

On Mon, Nov 5, 2012 at 12:17 PM, Prakash Singh wrote:
> Maarten, thanks a lot for the quick reply. I agree to your words but I
> was told that this is it so I thought of clarifying myself.
>
> the estimated model includes
> log of sector gdp (dependent variable)
> log of credit allocated (independent variable)
>
> and I have used Panel cointegration technique.
>
> Please let me know if more information is required
>
>
> Regards
> Prakash
>
> On Mon, Nov 5, 2012 at 4:37 PM, Maarten Buis <maartenlbuis@gmail.com> wrote:
>> On Mon, Nov 5, 2012 at 11:54 AM, Prakash Singh wrote:
>>> Suppose I run regression for two different time period for same set of
>>> variables and get 0.13 and .34 (both statistically significant) as
>>> estimated coefficient of a variable. Now I want to know that: does
>>> this high coefficient value in the second period indicate improvement
>>> in the effect of the variable on the dependent variable.
>>
>> Possible, but not certain.
>>
>> 1) We don't know what the unit of either the dependent and independent
>> variable is, so we have no way of judging whether this is a
>> substantively meaningful or a substantively meaningless change.
>>
>> 2) You did not say which regression model you used. If it is a
>> non-linear model the comparison between periods becomes much harder,
>> some would say impossible.
>>
>> 3) We cannot say with the information you have given use whether we
>> can reject the hypothesis that these coefficients are equal. My
>> default way to perform such a test is to estimate one model and add
>> interaction terms, others prefer -suest-.
>>
>> -- Maarten
>>
>> ---------------------------------
>> Maarten L. Buis
>> WZB
>> Reichpietschufer 50
>> 10785 Berlin
>> Germany
>>
>> http://www.maartenbuis.nl
>> ---------------------------------
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/



-- 
---------------------------------
Maarten L. Buis
WZB
Reichpietschufer 50
10785 Berlin
Germany

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


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