Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Standard Errors for Skewness & Kurtosis


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Standard Errors for Skewness & Kurtosis
Date   Fri, 10 Sep 2010 18:58:57 -0400

As in:

**************
sysuse auto, clear
bootstrap r(skewness) r(kurtosis) , reps(10000): sum mpg, det
*************
I'm curious: why do you need these quantities?

Steve

On Fri, Sep 10, 2010 at 6:09 PM, Steve Samuels <sjsamuels@gmail.com> wrote:
> The exact standard deviations (and hence, standard errors) for any
> distribution, not just the Gaussian, can be found in either edition of
> CR Rao, Linear Statistical Inference and its Applications, Wiley.
> However  I suggest that you get the standard errors via jackknife or
> bootstrap.
>
> Steve
>
> Steven J. Samuels
> sjsamuels@gmail.com
> 18 Cantine's Island
> Saugerties NY 12477
> USA
> Voice: 845-246-0774
> Fax:    206-202-4783
>
> On Fri, Sep 10, 2010 at 9:52 AM, Stas Kolenikov <skolenik@gmail.com> wrote:
>> Check the code; it must compute the standard errors somehow, although
>> it might do so under the assumption of normality. At any rate, the
>> standard error for kurtosis will be a function of the eighth moment of
>> the data, and that's a lot to ask: it may not exist, and even if it
>> does, it will be quite unstable unless you have a few thousand
>> observations.
>>
>> If Stata had contour plots, you could've constructed a cute empirical
>> likelihood diagram for the joint confidence set of skewness and
>> kurtosis (I've got the basic empirical likelihood code, but I've no
>> clue as to how to proceed with contour plots; I know Sergiy Radyakin
>> did some of them, but I am not sure mere mortals can reproduce his
>> work in Stata graphics :)). I think you can do this in R, I vaguely
>> remember doing this myself.
>>
>> On Thu, Sep 9, 2010 at 10:19 PM, Philip Burgess
>> <philip.burgess.uq@gmail.com> wrote:
>>> Stas;
>>>
>>> No - I've checked the Saved Results for sktest with no joy.
>>>
>>> Thanks;
>>>
>>> Philip
>>>
>>> On Fri, Sep 10, 2010 at 1:08 PM, Stas Kolenikov <skolenik@gmail.com> wrote:
>>>> On Thu, Sep 9, 2010 at 9:52 PM, Philip Burgess
>>>> <philip.burgess.uq@gmail.com> wrote:
>>>>> I can't find any ado that calculates the standard errors for skewness
>>>>>
>>>>> Do you know of such or how I might estimate these?
>>>>
>>>> Is there anything available as a by-product of -sktest-? If not, the
>>>> ado-file must compute the standard errors to perform the test, I
>>>> believe.
>>>>
>>>> --
>>>> Stas Kolenikov, also found at http://stas.kolenikov.name
>>>> Small print: I use this email account for mailing lists only.
>>>> *
>>>> *   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/
>>>
>>
>>
>>
>> --
>> Stas Kolenikov, also found at http://stas.kolenikov.name
>> Small print: I use this email account for mailing lists only.
>>
>> *
>> *   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   |   Site index