When given a mean and variance of a sample, without knowing the observations, how would you then find the new mean and variance given more observations?
Any help with this would be much appreciated.
Thank you in advance
When given a mean and variance of a sample, without knowing the observations, how would you then find the new mean and variance given more observations?
Any help with this would be much appreciated.
Thank you in advance
Let us assume that $\left( x_1, \ldots, x_n \right)$ is the sample and let us call $\mu_n$ the empirical mean of size $n$. Then the updating rule is given by $\mu_n = \frac{n - 1}{n} \mu_{n - 1} + \frac{1}{n} x_n$. This is how to derive the result \begin{eqnarray*} \mu_n & = & \frac{1}{n} \sum_{i = 1}^n x_i\\ & = & \frac{n - 1}{n} \times \frac{1}{n - 1} \sum_{i = 1}^{n - 1} x_i + \frac{1}{n} x_n\\ & = & \frac{n - 1}{n} \mu_{n - 1} + \frac{1}{n} x_n \end{eqnarray*} The rule for the variance is a bit more complicated but follows from the same reasoning.