This question is about R programming. I am not sure what to do with the question.

Could you help me?

Thanks,

(Multivariate normal). Multivariate normal distribution can be derived by linear combination of independent normal distributions. This property can also be used to generate dependent normal random vectors. In particular, let 11 a two dimension vector and E be a 2 x 2 positive denite matrix. To derive a 2-dimensional normal random vector with expected vector equal to p, and variance-covariance matrix equal to 2, you can (i) Calculate eigen decomposition of 2 such that it is decomposed as 2: = UDUT, where U is an orthogonal matrix and D is a diagonal matrix. Let 21/2 : UD1/2UT. (ii) Generate X1, X2 independently from N(0, 1). Let a: 2 (X1, X2)T. (iii) Let y = p, | 21/ 2x. Then, y is the twodimensional random vector with expected vector p, and variancecovariance matrix 2. Please use this method to generate y\"Nl(3:i)w(i ill- To check whether your method is correct, you need to independently generate n = 105 data. Let them be denoted by y1,y2,- - - ,yn. Then, you calculate the sample mean (which should be a 2-dimensional vector), and their sample variance-covariance (which should be a 2 X 2 matrix) of the data. Compare those with p, and E and draw your conclusion. You also need to include histogram plot to check it looks like normal