Constructing an upper confidence limit for $σ^2$

33 Views Asked by At

Suppose that $X_1, ..., X_n$ form a random sample from the normal distribution with unknown mean µ and unknown standard deviation σ.

Construct an upper confidence limit U(X1, ..., Xn) for $σ^2$ such that

$P[σ^2 < U(X_1, ..., X_n)]= 0.99$.

What I've done is so far

Since $\dfrac {(n-1)s^2}{σ^2}$~ $X^2(n-1)$,

$P(X^2_.99 <\dfrac {(n-1)s^2}{σ^2})$=$P(σ^2<\dfrac {(n-1)s^2}{X^2_.99})$

Therefore $U(X_1, ..., X_n)=\dfrac {(n-1)s^2}{X^2_.99}$

Now is it correctly done?