1

Suppose I have two data samples of the same size N generated from two separate distributions, and a new number is given. How could I find out which distribution this new number is coming from? What should I do if the sample size were different?

My first thought is to compare the empirical probabilities based on these two samples, but I am not quite sure if it is the optimal solution. I am very grateful for any help.

www
  • 43

1 Answers1

1

Given $x$, we compare $P(D_1|x)$ and $P(D_2|x)$ and decide which has a higher value.

To compute these values though, some assumptions and prior knowledge are required.

From Bayes' rule, this is equivalent to comparing $P(D_1)P(x|D_1)$ and $P(D_2)P(x|D_2)$. We have to make assumption about $P(D_i)$.

Siong Thye Goh
  • 149,520
  • 20
  • 88
  • 149