Services & ResourcesWolfram ForumsMathGroup Archive

[Date Index] [Thread Index] [Author Index]

Re: Re: Re: Multivariate-T Simulation Problem

• To: mathgroup at smc.vnet.net
• Subject: [mg39032] Re: [mg39018] Re: Re: [mg38986] Multivariate-T Simulation Problem
• From: Dr Bob <drbob at bigfoot.com>
• Date: Sat, 25 Jan 2003 01:24:44 -0500 (EST)
• References: <200301241008.FAA04684@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

It's not rejected at ALL confidence levels --- just the ones with alpha > 
10^(-4000).

Nevertheless, that's good evidence of a bug.

Bobby

On Fri, 24 Jan 2003 05:08:06 -0500 (EST), Kyriakos Chourdakis 
<tuxedomoon at yahoo.com> wrote:

> A variance test gives p-values of order 10^(-4000)
> which apparently reject the null at all confidence
> levels. I am pretty certain it is bug too.
>
> K.
>
> <<============>
> You might want to use VarianceTest to give a measure
> of how far off the sample variance is.  Without
> something like this, "the sample variance is wrong"
> doesn't mean anything.
>
> __________________________________________________
> Do You Yahoo!?
> Everything you'll ever need on one web page
> from News and Sport to Email and Music Charts
> http://uk.my.yahoo.com
>
>



-- 
majort at cox-internet.com
Bobby R. Treat

• References:
  • Re: Re: Multivariate-T Simulation Problem
    • From: Kyriakos Chourdakis <tuxedomoon@yahoo.com>