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: How to predict in mixed level linear regression when key predictor variable is previous years result
From
Yuval Arbel <[email protected]>
To
[email protected]
Subject
Re: st: How to predict in mixed level linear regression when key predictor variable is previous years result
Date
Sun, 27 Nov 2011 02:15:12 -0800
Paul,
If the estimated b3 in your model is between -1 and 1 you might be
able to calculate the long-run multiplier. I suggest you take a look
at econometric textbooks that deal with lagged dependent variables and
particularly the geometric-lag model.You should also take a look at
VAR models and Impulse-Response functions - I believe this is what you
need.
The literature on this subject of time-series analysis is growing at
exponential rate.
If you are somewhat familar with matrix algebra you may take a look at:
Johnston and Dinardo, Econometric Methods, 4th Edition: Chapter 9 -
starting from p. 287. In addition, in Appendix A, starting from p. 455
you have a repitition on matrix algebra and particularly look on pages
476-484. This material is necessary to understand the Johansen
coitegration test of the VAR model
Here are also two of my papers in which I use the VAR model (in the
first part of the paper) and geometrical-lag model:
Arbel, Yuval; Ben Shahar,Danny; Gabriel, Stuart and Yossef Tobol:
"The Local Cost of Terror: Effects of the Second Palestinian Intifada
on Jerusalem House Prices".Regional Science and Urban Economics (2010)
40: 415-426
Arbel, Yuval; Ben Shahar,Danny; and Eyal Sulganik: "Mean Reversion and
Momentum: Another Look at the Price-Volume Correlation in Real Estate
Markets" Journal of Real Estate Finance and Economics (2009) 39:
316-335
In addition, if you have several countries - you might want to make a
panel data analysis. Again, you should take a look at econometric
textbooks. Again, a subject with a literature that grows at
exponential rate, and here you can also take a look at Johnston and
Dinardo
On 11/27/11, Hunter Paul Prof (MED) <[email protected]> wrote:
> Dear all
>
> I am trying to predict the results of a mixed level linear regression model
> of time series
>
> I have time series data from several countries and have developed a mixed
> level linear regression model with country as the level variable as follows
>
> Z= a+ b1*Year +b2*X+b3*Z(in previous year)
>
> Now I can use
>
> predict Pred_X, fitted level( country)
>
> to get good predictions of Z up to the year after I have measured data for Z
> so have Z(in previous year). How can I predict further into the future?
>
> Best Wishes
>
> Paul
>
>
>
> *
> * 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/