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