chi2bintest (Scripts) Publisher's description
from Peder Axensten
Given a number of samples with a binary outcome ("is or is not something")
Usage: [p, Q]= chi2bintest(x, n)
The chi-squared binary test.
Given a number of samples with a binary outcome ("is or is not something"),
this function tests the hypothesis that the samples are independent.
If Q > chi2(p, nu), the hypothesis is rejected.
If x is a vector, each item is the result of a sample.
n must be a scalar (same total for all samples) or a vector of the same length as x.
If x is a matrix, each row is the result of a sample. Each column is an independent
sample series. n must be a scalar (same total for all samples) or
a vector of the same length as the number of samples in x (same totals for all sample
series) or a matrix of the same size as x.
If you find any errors, please let me know: .
x Relative frequencies of the outcome, if 0<=x<=1. Absolut number count, if integer x>=0.
n Number of tries.
p The prob ability value, calculated from Q.
Q The resulting Q-value.
In region A, 324 of 556 cows were red, whereas in region B 98 of 260 were red.
[p, Q]= chi2bintest([324; 98], [556; 260])
With an error risk of about 4e-08, we can claim that the samples are independent.
In three regions, reindeers were cheched for four various deseases.
p= chi2bintest([28,39,8,4782; 17,7,5,903; 21,14,6,2322], [15996; 10127; 12476])
p = 0.9869 0.0007 0.9973 0
It seems that the first and third deseases are region independent, whereas the others
have regional dependencies (are independent to each other).
HISTORY: v.1.0, first working version, 2007-08-30.
COPYRIGHT: (c) 2007 Peder Axensten. Use at own risk.
System Requirements:MATLAB 7.4 (R2007a)
Program Release Status: New Release
Program Install Support: Install and Uninstall