Chi-Square Statistic for Counts

Problem

Return the chi-square statistic sum((o - e)^2 / e) for a 2D table of observed counts, where e = row_total * col_total / total. Return 0.0 for an empty table, a zero total, or any zero expected count.

Hint

Compute the row and column totals once, then loop over the cells.

Test cases

case 1

in: [[[10,20],[30,40]]]

expected: 0.7936507936507936

case 2

in: [[[5,5],[5,5]]]

expected: 0

case 3

in: [[[1,2,3],[3,2,1]]]

expected: 2

case 4

in: [[[10,0],[0,10]]]

expected: 20

case 5

in: [[[10,20,30]]]

expected: 0

Starter codePython

def chi_square_counts(observed):
    # Your code here
    pass

5 test cases · run your code against them in the editor