help-octave
[Top][All Lists]
Advanced

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

Re: Problems with Statistics package


From: marco atzeri
Subject: Re: Problems with Statistics package
Date: Wed, 30 May 2012 08:58:08 +0200
User-agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:12.0) Gecko/20120428 Thunderbird/12.0.1

On 5/30/2012 7:56 AM, Hannes wrote:
Hi everyone,

I am using the octave statistics package right now, but have some
unexpected (possibly bugged) results.

I have attached a log that shows the problem. On a perfect grouping,
anova produces p-val of 1 instead of 0. The f-statistic being displayed as
-94353808874586048.0000
leads me to the conclusion that some internal number limits where
exceeded, which is strange, because the example is very simple, the
involved numbers very low etc.

Could someone help me with that?

Thank you very much!

Best,
Hannes

PS: System is:

octave:83> version
ans = 3.2.4
octave:84> uname
ans =
{
sysname = Linux
nodename = research-macpro01
release = 3.2.0-24-generic
version = #38-Ubuntu SMP Tue May 1 16:18:50 UTC 2012
machine = x86_64
}



There is an overflow:

octave:4> anova(samples,grps)
warning: division by zero

One-way ANOVA Table:

Source of Variation   Sum of Squares    df  Empirical Var
*********************************************************
Between Groups              944.4969     1       944.4969
Within Groups                 0.0000   159         0.0000
---------------------------------------------------------
Total                       944.4969   160

Test Statistic f                 Inf
p-value                       0.0000

ans = 0




octave:5> version
ans = 3.6.1
octave:6> uname
ans =

  scalar structure containing the fields:

    sysname = CYGWIN_NT-6.1-WOW64
    nodename = MARCOATZERI
    release = 1.7.16s(0.261/5/3)
    version = 20120525 17:01:08
    machine = i686


with statistics-1.1.0


Regards
Marco


reply via email to

[Prev in Thread] Current Thread [Next in Thread]