CFI Lite

Problem

Return the comparative fit index 1 - max(chi2 - df, 0)/max(chi2_null - df_null, 0), clamped to [0, 1]. Return 0.0 when a degrees-of-freedom value is not positive or the null denominator is not positive.

Hint

CFI compares the fitted model against a baseline independence model.

Test cases

case 1

in: [30,10,200,15]

expected: 0.8918918918918919

case 2

in: [5,10,200,15]

expected: 1

case 3

in: [50,20,60,25]

expected: 0.1428571428571429

case 4

in: [30,0,200,15]

expected: 0

case 5

in: [30,10,10,15]

expected: 0

Starter codePython

def cfi_lite(chi2, df, chi2_null, df_null):
    # Your code here
    pass

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