variance of a Maximum Liklihood estimator

47 Views Asked by At

I am reading a book on Bayesian Estimation and Sensor Fusion and I want to know where the formula below come from. In fact, what is the relation between the variance and the second derivative of the ML estimation?

enter image description here

enter image description here

1

There are 1 best solutions below

1
On BEST ANSWER

It is the variance of a weighted sum of independent RVs (in a slightly odd form)

If: $$ \hat{x}=\sum_{k=1}^n \alpha_k x_k $$

where the $X_k, k=1\dots n$ are independent RVs then:

$$ \mbox{Var}(\hat{X})=\sum_{k=1}^n \alpha_k^2 \mbox{Var}(X_k) $$

But: $\alpha_k=\frac{\partial \hat{x}}{\partial x_k}$