Message Boards Message Boards

Get right results using FindDistribution - "TargetFunctions"?

Dear all,

I would like to get the AIC and BIC for several distributions relating to an empirical data set. As I don't want to get the best fitting once (I always get "MixtureDistribution" as a result, which is not really satisfying) I tried to get the specific results for several distributions. Unfortunately, after using this code:

FindDistribution[data, 
 TargetFunctions -> {ParetoDistribution, NormalDistribution, 
   LaplaceDistribution}, {"AIC", "BIC"}]

I just get this as a result:

SymbolicMachineLearning`file19FindDistribution`PackagePrivate`\
iFindDistribution[{0.00293084, 0.0129467, -0.00500067, -0.00462811, 
  0.0059465, -0.0199607, -Log[372]

Does anyone know whats the problem here? I don't get what it wants to tell me with: "SymbolicMachineLearningfile19FindDistributionPackagePrivate`\iFindDistribution"

Thank you in advance!

Best regards Alex
POSTED BY: Alexander Hempfing
2 Replies
Sort By:
Jim Baldwin
Jim Baldwin, Retired
Posted 7 years ago

You need to rearrange the parameters for FindDistribution and give the number of "best" distributions to display:

data = RandomVariate[LaplaceDistribution[3, 4], 100];
FindDistribution[data, 3, {"AIC", "BIC"}, TargetFunctions -> {ParetoDistribution, NormalDistribution, LaplaceDistribution}]

{{LaplaceDistribution[2.2087, 4.33898], {-6.17796, -6.19207}}, {NormalDistribution[2.2087, 6.81179], {-6.47878, -6.49288}}, {ParetoDistribution[351.35, 34.1949, 0.735711, -23.8784], {-8.28244, -8.30892}}}
POSTED BY: Jim Baldwin

Attaching data file would potentially get you a bit more attention.
POSTED BY: Sam Carrettie
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Tag limit exceeded
Note: Only the first five people you tag will receive an email notification; the other tagged names will appear as links to their profiles.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract