Re: Gnumeric regression is wrong?





On Thursday 27 September 2001 06:36 pm, you wrote:
There is definitely something wrong with the regression calculations in
the data analysis tools. It seems that the standrard error for the
parameter estimates are switched (and then all the calculations based on
them seem to be wrong).

Yes, I just observed that too. However, for something like quadratic fit 
regression, the error seems to be more severe. 
The following I attach a file, that I originally created using MS Excel. I 
opened it using gnumeric, and copy the data, and did Regression in the second 
sheet (I put label in the title for which one was created using Excel and 
which one was created using gnumeric, so it should be clear.)

What I did basically I select Data Analysis, Tool, Regression, put the 
dependent data in Input Y Range, and the independent data in Input X Range, 
and hit OK.

Let me know if I can be of any more help.

Reuben D. Budiardja

Attachment: age_data.gnumeric
Description: GNU Zip compressed data

Attachment: hogs_data.gnumeric
Description: GNU Zip compressed data



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