For this question, you have to simulate some datasets. Let$$s assume the outcome Y depends$$linearly on $40$ covariates X$1,$X$2,...,$X$40.$ The relationship is given by the following equation:$$Y $=\backslash$beta $0+\backslash$beta $1$X$1+\backslash$beta $2$X$2++\backslash$beta $40$X$40+(1)($a$)[8$ Marks$]$ Perform the following simulations.$$ Generate training sets of size $50$ as follows:$$ Generate $50$ random values for all X variables from a standard normal distribution.$$That is$,$ X$1$ N$(0,1),$X$2$ N$(0,1),...,$X$40$ N$(0,1).$ Generate $$ also from a standard normal distribution: $$ N$(0,1).$ Generate $\backslash$beta values from a Uniform distribution where $\backslash$beta $1$ to $\backslash$beta $20$ are simulated$$from Uniform$(0\mathrm{.}5,1\mathrm{.}5)$ and $\backslash$beta $21$ to $\backslash$beta $40$ are simulated from Uniform$(0\mathrm{.}1,0\mathrm{.}4).$ You$$can choose any arbitrary value for $\backslash$beta $0.$ Then generate Y using $(1).$Note: You cannot use a for loop for this data generation. If you use a for loop,$$you will receive a zero for this section.$$ Generate a test set of size $1000$ as follows:$$ Generate $1000$ random values for all X variables from a standard normal distribution.$$That is$,$ X$1$ N$(0,1),$X$2$ N$(0,1),...,$X$40$ N$(0,1).$ Generate $$ also from a standard normal distribution: $$ N$(0,1).$ Use the same $\backslash$beta values generated for the training set.$$ Then generate Y using $(1).$Note: You cannot use a for loop for this data generation. If you use a for loop,$$you will receive a zero for this section.