Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: ksmirnov
From
Nick Cox <[email protected]>
To
[email protected]
Subject
Re: st: ksmirnov
Date
Mon, 20 Aug 2012 17:40:10 +0100
I am disinclined to support what I guess is likely to be wrong in
principle, but -help density functions- guides you on functions for t
distributions.
Nick
On Mon, Aug 20, 2012 at 5:16 PM, Guessab Houda
<[email protected]> wrote:
> Thank you nick,
>
> Apologies for posting anonymous question... this is such a good idea to make quantile quantile plot. The thing is that I have to use kormogolov smirnov but I really don't know how to apply it to a distribution with t with 8 d.f.Could you please help me?
>
>
> Houda GUESSAB
>
>
>
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Nick Cox
> Sent: lundi 20 août 2012 17:40
> To: [email protected]
> Subject: Re: st: ksmirnov
>
> Naturally I agree with Maarten on the point of list practice. He is
> referring to an explicit request that posters use their full names.
> 2.1.3 within http://www.stata.com/support/faqs/resources/statalist-faq/#tojoin
> means exactly what it says.
>
> On the substantive question,
>
> (1) if "series" means what I think it does, a time series of some
> kind, then note that Kolmogorov-Smirnov tests tacitly assume mutual
> independence
>
> (2) the most direct way of comparing a distribution with t with 8 d.f.
> is a quantile-quantile plot.
>
> Example code to show how easy this is:
>
> sysuse auto, clear
> egen rank = rank(mpg), unique
> su mpg, meanonly
> local N = r(N)
> gen tquantile = -invttail(8, (rank - 0.5) / `N')
> scatter mpg tquantile
>
>
> On Mon, Aug 20, 2012 at 4:17 PM, Maarten Buis <[email protected]> wrote:
>> On Mon, Aug 20, 2012 at 5:05 PM, houda wrote:
>>> I m trying to find a simple way to test if a series is t-distributed with 8
>>> degrees
>>> of freedom using a Kolmogorov-Smirnov test, could you please help me?
>>
>> Policy on the statalist is not to post anonymous questions. This is
>> clearly explained in the Statalist FAQ.
*
* 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/