commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Phil Steitz <>
Subject Re: [math] API changes for RC2
Date Sun, 26 Sep 2004 22:58:27 GMT
Kim van der Linde wrote:
> Al Chou wrote:
>> -0; question:  do people really use the population version?  When 
>> dealing with
>> real-world data or any other distribution that is not known a priori, 
>> how could
>> you, in good conscience?  And if you know the distribution a priori, 
>> why would
>> you need to compute statistics about it?
> It depends if you want to minimise error or bias. 

This is a good point. Strike the "haughty" assertion in my last post -- 
i.e. it is not always "bogus" in practical terms to use a biased estimator.

> We do use it at times, 
> just like we do change degrees-of-freedom (in many of our multivariate 
> analyses by 4 to be precise due to three modifications that take 1, 1, 
> and 2 degrees of freedom) in all kind of tests to correct for previous 
> calculations on the same data set. In some studies, the variance is 
> calculated relative to a set average, which become biased 
> the-other-way-around if you would use sample variances.

Yes, this was the first case that I mentioned.
> Anyway, I reported these examples back because of the real life usage.
> Cheers,
> Kim

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message