Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: xtdpdsys/xtabond2 and xtdpd command Question

From   Nick Cox <>
Subject   Re: st: xtdpdsys/xtabond2 and xtdpd command Question
Date   Wed, 9 May 2012 22:04:54 +0100

OK, but my second point was to -summarize- on the same variables.

The -cmdname- convention, explained in the FAQ, is that -cmdname-
means the Stata command with that name.

The commands you are trying are presumably rejecting your data because
of either (a) string variables or (b) missing values. So -summarize-
is a way of trying to find out what is going on. -summarize- on a
string variable shows 0 observations. Missing values on a numeric
variable will show up as a reduced number of non-missing values.

On Wed, May 9, 2012 at 9:58 PM, Rituparna Basu <> wrote:
> Sorry about that Nick.
> The panel data set that I have, has 467 groups with 27 months as mean length of follow-up. The variable, say 'x', I am referring to is continuous and is dynamic/ autoregressive in nature. So I tried the dynamic panel models.
> Xtabond did not seem to work may be because the dataset is unbalanced in nature. Next, I tried xtdpdsys and xtdpd model, but they said 'no observations'. I believe, I am going wrong in specifying the model.
> I am interested in running a model that uses at least two lags of the variable. Currently, I have no independent variables. Let me know if more clarification is needed.
> RB
> -----Original Message-----
> From: [] On Behalf Of Nick Cox
> Sent: Wednesday, May 09, 2012 1:28 PM
> To:
> Subject: Re: st: xtdpdsys/xtabond2 and xtdpd command Question
> Try _not shouting please_ and -summarize- on the same variables.
> Nick
> On Wed, May 9, 2012 at 8:27 PM, Rituparna Basu <> wrote:
>> I have an unbalanced panel data in which I am trying to run the 'xtdpdsys' and 'xtdpd' commands. But it gives me error saying NO OBSERVATIONS. I am not sure where I am going wrong. Currently, I am running an INTERCEPT only model.
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index