Re: st: Re: xtunitroot ips error for 'no observations' and trend

[Scott Merryman <scott.merryman@gmail.com>]

Without trend, failure to reject the null of a unit root would imply
that the data needs to be differenced to be stationary; if the null is
rejected then the data is stationary and doesn’t need to be
differenced.

With trend, failure to reject implies that the data needs to be
differenced to make it stationary ; if the null is not rejected the
data is trend stationary and needs to be analyzed a time trend in the
regression model instead of differencing.

Scott


On Sun, Sep 19, 2010 at 6:56 PM, wangpan110 <wangpan110@hotmail.com> wrote:
> Hi Scott, 1) does the involvement of -trend- mean the unit root test account
> both DSP and TSP?
>             2) the involvement of trend changes the p-value substantially,
> does it mean the nonstationarity of this panel/series is closely related
> with TSP? if it is, what i should do?
>
> many thanks
>
> Pan

*
*   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/