Let A = 5 Consider the population A , A+2 ,A+4 , A+6 , A+8 Draw all posible sample of size 3 without replacement.If A is odd number verily the biasedness property of sample variance S^2
Samples of size 3 without replacement:
{5,7,9}; {5,7,11}; {5,7,13}; {5,9,11};{5,9,13}; {5,11,13}; {7,9,11}; {7,9,13}; {7,11,13}; {9,11,13}
If A is odd number verily the biasedness property of sample variance S2.
"S^2_n=\\frac{1}{n(n-1)}\\sum_{i\\neq j} \\frac{(X_i-X_j)^2}{2} = \\frac{1}{n(n-1)}((n-1)\\sum_{i=1}^n X_i^2-\\sum_{i\\neq j} X_iX_j)"
The expectation of sample variance:
"E[S^2_n]=\\frac{\\sum_{i=1}^n E[X_i^2]}{n}-\\frac{\\sum_{i\\neq j} E[X_i*X_j]}{n(n-1)}"
Since we have independent variables:
"E[X_iX_j]=E[X_i]*E[X_j]"
Let assume that:
"E[X_i^2]=\\mu_2" and "E[X_i]=\\mu_1"
Then:
"\\sum_{i=1}^n E[X_i^2]=n*\\mu_2" and "\\sum_{i\\neq j} E[X_i]E[X_j]=n(n-1)\\mu_1^2"
Finally:
"E[S_n^2]=\\mu_2-\\mu_1^2=Var[X]"
So, S2 is unbiased.
Comments
Leave a comment