Statalist


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

Re: st: regression on variable over time to get mean autoregressive parameter


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: regression on variable over time to get mean autoregressive parameter
Date   Tue, 7 Oct 2008 18:23:18 -0400

Fabian Brenner <Fabian.Brenner@gmx.de>:
Try
g id=group(bacteria_name)
tsset id year
reg number l.number

but you might want a log-log specification:
g lnn=ln(number)
reg lnn l.lnn


On Tue, Oct 7, 2008 at 4:24 PM, Fabian Brenner <Fabian.Brenner@gmx.de> wrote:
> Dear all:
>
> I have observations for different years ("year" 1990-2000) for bacterias ("bacteria_name", eg.ATT, GLY, ...) and the number of the bacteria population ("number").
>
> Now I would like to run the following regression: number (year+1) = w0 + w1 * number (year) + e for every two consecutive years (whenever there are values for at least two consecutive years for a given bacteria_name) to get mean estimates of the parameters w0 and w1 (w1 should reflect the fade rate of the bacteria population).
>
> I hope someone has a solution or can give me a hint to solve this problem. I tried it many times but I did not found a solution (especially how to get the mean of the different regressions and on how to refer to the last observation)..Thanks a lot.
>
> Fabian Brenner
>
> My data looks like the following (using -sort bacteria_name year-)
> bacteria_name        year         number
> ATT                    1990           10
> ATT                    1991             3
> GLY                    1993             4
> SIS                    1992           99
> SIS                    1993           19
> SIS                    1994             8
> X                       1995             2
> .
*
*   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   |   What's new   |   Site index