Statalist


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

RE: RE: st: ARIMA model errors


From   "Erlinger, Thomas P." <TErlinger@seton.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: RE: st: ARIMA model errors
Date   Thu, 8 Oct 2009 10:13:35 -0500

Well, for some models I find that it won't generate values beyond the
next observation even after tsappend unless I add the dynamic option.
Also, in the example I gave below, there is an added variable "diff2"
that is included in the model.  Is this an issue?  Still don't know how
to generate error estimates after either method. 

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Robert A
Yaffee
Sent: Thursday, October 08, 2009 10:06 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: RE: st: ARIMA model errors

Tom,
   Your syntax looks fine but why are you using the dynamic option?
I would prefer the one-step ahead, as the forecast errors are not so
large.
  - Bob

Robert A. Yaffee, Ph.D.
Research Professor
Silver School of Social Work
New York University

Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf

CV:  http://homepages.nyu.edu/~ray1/vita.pdf

----- Original Message -----
From: "Erlinger, Thomas P." <TErlinger@seton.org>
Date: Thursday, October 8, 2009 10:59 am
Subject: RE: st: ARIMA model errors
To: statalist@hsphsun2.harvard.edu


> this is what I run after a tsappend and it generates undifferenced 
> values.  Is this incorrect?
> 
> arima s365.encounters in 2740/3568, ar(1 6) ma(1) predict yhat1 in 
> 2740/3573, y predict res1 in 2740/3573, yres
> 
> arima s365.encounters diff2 in 2738/3568, ar(1 6) ma(1) predict yhat5 
> in 2740/3573,  dynamic(d(08oct2009)) y predict res5 in 2740/3573, yres
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Robert A 
> Yaffee
> Sent: Thursday, October 08, 2009 9:52 AM
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: ARIMA model errors
> 
> Tom,
> When I test the y option with a differenced series, it does not 
> generate an undifferenced forecast.
>     Bob
> 
> 
> Robert A. Yaffee, Ph.D.
> Research Professor
> Silver School of Social Work
> New York University
> 
> Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf
> 
> CV:  http://homepages.nyu.edu/~ray1/vita.pdf
> 
> ----- Original Message -----
> From: "Erlinger, Thomas P." <TErlinger@seton.org>
> Date: Thursday, October 8, 2009 10:37 am
> Subject: st: ARIMA model errors
> To: statalist@hsphsun2.harvard.edu
> 
> 
> > Is there a way to get error, prediction interval, confidence
intervals
> > from ARIMA forecasts of differenced data?   Using the "y" option
gets
> > predictions and residuals, but I can't figure out how to get the
> other.
> > 
> > Any help appreciated.
> > 
> > Tate
> > 
> > *
> > *   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/
> *
> *   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/
> 
> *
> *   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/
*
*   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/

*
*   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