After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 633735 - Exponential fit in graphs is flawed for small values.
Exponential fit in graphs is flawed for small values.
Status: RESOLVED FIXED
Product: libgoffice
Classification: Other
Component: Graphing / Charting
unspecified
Other Linux
: Normal normal
: ---
Assigned To: Jean Bréfort
Jody Goldberg
Depends on:
Blocks:
 
 
Reported: 2010-11-01 19:58 UTC by Jean Bréfort
Modified: 2010-11-01 20:06 UTC
See Also:
GNOME target: ---
GNOME version: ---



Description Jean Bréfort 2010-11-01 19:58:11 UTC
Originally reported as https://bugs.launchpad.net/ubuntu/+source/libio-digest-perl/+bug/599534.

The exponential fit is broken in graphs. Using the following data plot
A B
0.000568 0.22
0.001136 0.05
0.00170 0.02
0.000136 0.97
0.000329 0.42

Plot and add a trend-line. Make the trend-line exponential and you get a garbage fit. If you multiply column A by 4, the fit works. Multiply by 3 and it fails. It appears that the algorithm used for this type of trend-line is flawed and has limitations that do not occur in MS Excel.

Actually, logest returns 0 as m1 coefficient because the actual result is too small to be represented as a double. The solution is to have a regression procedure returning the logarithms of the coefficients instead of the coefficients.
Comment 1 Jean Bréfort 2010-11-01 20:06:08 UTC
This problem has been fixed in our software repository. The fix will go into the next software release. Thank you for your bug report.