Dogggg / accord

Automatically exported from code.google.com/p/accord
0 stars 0 forks source link

Binomial Test Incorrect P-Value Two-Tailed #40

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago
The Binomial test yields an incorrect P-Value for the Binomial Two-Tailed test 
but is correct for both One Tailed results compared to SciPy and R.

What steps will reproduce the problem?

1. use Constructor: BinomialTest(int successes, int trials, double    
hypothesizedProbability, OneSampleHypothesis alternate);
2. No. of trials 235
3. No. of successes 51
4. HypothesizedProbability 1/6

What is the expected output? What do you see instead?
Two Tailed
Statistic = 51
Accord.Net = 1.96404518910672
R = 0.04375
SciPy= 0.043747970182413443

What version of the product are you using? On what operating system?

v2.8.2

Please provide any additional information below.

Original issue reported on code.google.com by andrewhd...@gmail.com on 28 Feb 2013 at 3:26

GoogleCodeExporter commented 8 years ago
Thanks, will be fixed on next release. Binomial was being assumed symmetric in 
cases where it shouldn't.

Original comment by cesarso...@gmail.com on 4 Mar 2013 at 2:48

GoogleCodeExporter commented 8 years ago
Fixed on 2.9.0.

Original comment by cesarso...@gmail.com on 20 Apr 2013 at 5:25