estimation of $\mu$ in a Gaussian with set confidence interval

45 Views Asked by At

What is the step by step solution for solving the given problem? Searching the Web I am overwhelmed with many formulas and less illustrative examples.

Given a Gaussian distribution with unknown mean $\mu$ and variance=1, we want to estimate $\mu$ with an accuracy of epsilon and confidence interval of ci. What is the minimum number of iid samples we would need?

1

There are 1 best solutions below

2
On

The number of samples you need depends on your estimator.

For example, the maximum likelihood estimator of the mean of a Gaussian is the sample average: $\frac{1}{n} \sum_{i=1}^n x_i$. Note that the sample average for a sample of $n$ i.i.d. $N(\mu,1)$ random variables follows a $N(\mu,\frac{1}{n})$ distribution.

Using this, you can find the probability that the estimate is in the interval $(\mu-\epsilon, \mu+\epsilon)$ as a function of the number of samples $n$ by using the cumulative distribution function of $N(\mu,\frac{1}{n})$ .