Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: getting around tsset time variable integerrestriction -Work with minutes???


From   David Greenberg <[email protected]>
To   [email protected]
Subject   Re: st: RE: getting around tsset time variable integerrestriction -Work with minutes???
Date   Tue, 07 Feb 2006 22:51:49 -0500

Are you sure that this is what you want? If you analyze all the
observations as a single series, you are asserting that the parameters
characterizing the series did not change over the duration of your
study. In some applications, that assusmption might be dubious. Why not
first break up your series into subseries short enough for Stata to
handle. Comparison of the models will then tell you whether this
assumption is valid or invalid. Another way to proceed would be to
aggregate your half-hours into larger time units, such as days. If you
don't need to model variation over the course of the day, that would
solve your problem. David Greenberg, Sociology Department, New York
University

----- Original Message ----- 
From: Zachary Byron Singer <[email protected]>
Date: Tuesday, February 7, 2006 10:23 pm
Subject: Re: st: RE: getting around tsset time variable integer
restriction -Work with minutes???

> The problem is not the units, it is the length of the series.  My 
> serieshas over 149,000 observations and tsset wants the time 
> variable stored as
> an integer, which limits me to a maximum of 65,000 observations since
> integers cannot be smaller than -32767 or greater than 32740.  Simply
> converting half hours to minutes doesn't solve the problem.  I need 
> a way
> to store my time variable as a long.
> 
> Zach
> 
> On Wed, 8 Feb 2006, Rajesh Tharyan wrote:
> 
> > Hi,
> >
> > Coudn' t you work with minutes??
> >
> > gen min=(original_half_hour_variable)*60
> > and then use the tsset command
> >
> > regards
> > rajesh
> >
> >
> > -----Original Message-----
> > From: [email protected]
> > [mailto:[email protected]] On Behalf Of 
> Zachary Byron
> > Singer
> > Sent: 08 February 2006 03:02
> > To: [email protected]
> > Subject: st: getting around tsset time variable integer restriction
> >
> > Hi,
> >
> > Does anyone know how to get around the integer limitation for 
> tsset?  I am
> > dealing with a very large time series: 149,000 observations, each
> > representing a different half hour from october 1996 to april 2005.
> > tsset will not allow me to use the half hour as my unit of time, 
> since the
> > time variable is restricted to integers, ie numbers between [-32767,
> > 32740].  If stata cannot handle this, is there another software 
> package> that can?
> >
> > Thanks,
> >
> > Zach Singer
> > Harvard College 2006
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/support/faqs/res/findit.html
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> >
> > *
> > *   For searches and help try:
> > *   http://www.stata.com/support/faqs/res/findit.html
> > *   http://www.stata.com/support/statalist/faq
> > *   http://www.ats.ucla.edu/stat/stata/
> >
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
> 
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index