# st: Re: Help for survial analysis,21st Aug

 From khan ali To statalist@hsphsun2.harvard.edu, n.j.cox@durham.ac.uk Subject st: Re: Help for survial analysis,21st Aug Date Tue, 21 Aug 2007 00:23:57 -0700 (PDT)

```Dear Nick Cox,
McQueen and Thorley (1994) test the bubbles in US
stock prices. The article is cited as, McQueen, G.,
Thorley, S., 1994.Bubbles, stock returns, and duration
dependence. Journal of Financial and Quantitative
Analysis 29, 379-401.
They contend that if prices contain bubbles, then
runs of positive returns will exhibit negative
duration dependence, i.e., an inverse relation exists
between the probability of a run ending and the length
of the run. They adopted log-logistic model for
testing the null hypothesis of no rational bubbles.
The null hypothesis of no bubbles implies that the
probability of a run ending is independent of  the
prior returns or that the returns are random.In terms
of the model, the null hypothesis of  no duration
dependence is that slope(beta=0) .The bubble
alternative suggests that the probability of a
positive run ending should decrease with the run
length or that the value of the slope parameter ,beta
is negative.
RunLength	RunCount	Sample Hazard

1	       95	             0.487
2	       52	             0.52
3	       26	             0.542
4	        9              	 0.409
5	        2	             0.154
6	        4	             0.364
7	        1	             0.143
8	        4	             0.667
10	        1	             0.50
17	        1	             1.00
Log-Logistic test

alpha(const)	             0.051

beta(slope)	                  -0.283
LRT of Null hypothesi          3.75
p-value		0.05

I have a sequence of runs,little confused while
estimating slope parameter and LRT by log-logistic,
weibull model in stata.
I am using foloowing commands
stset RunLength

streg RunLength, dist(llogistic) time
streg RunLength, dist(weibull) time

I appreciate any answer and help.

Thanks
Haque,Ali

--- Nick Cox <n.j.cox@durham.ac.uk> wrote:

> This was posted earlier, but got no response,
> perhaps because it isn't clear what you seek.
>
> You state that you are a "little confused",
>
> Nick
> n.j.cox@durham.ac.uk
>
> khan ali (a.k.a. ABDUL-HAQUE)
>
> > I am using parametric models for a survival
> analysis.
> > I have a sequence of runs and want to test
> duration
> > dependence, a technique developed by McQueen and
> > Thorley (1994) to test the bubbles in stock
> prices.
> > They contend that if prices contain bubbles, then
> runs
> > of positive returns will exhibit negative duration
> > dependence, i.e., an inverse relation exists
> between
> > the probability of a run ending and the length of
> the
> > run. They adopted log-logistic model for testing.
> > I am little confused while estimating
> log-logistic,
> > weibull model.
> > I am using following commands
> >
> > stset curun
> > streg curun, dist(llogistic) time
> > streg curun, dist(weibull) time
> >
>
> *
> *   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/
>

____________________________________________________________________________________
Boardwalk for \$500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games.
http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow

*
*   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/
```