Help with pattern recognition homework using matlab

The question is:

Since P(w1) = P(w2) = P(w3), the posterior probabilities P(wi|x); i = 1; 2; 3; are determined by the class-conditional densities p(x|wi); i = 1; 2; 3. Hence, we first estimate/train p(x|wi); i = 1; 2; 3 with MLE approach. Assume that p(x|wi)~N(MUi,SIGMAi)

Use the fi rst 1000 samples of each category to estimate p(x|wi) and plot the resulting class-conditional densities.

My question:

should i write a matlab code? how am i supposed to estimate p(x|wi)?

Thank you.