Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I am given two random variables $X$ and $Y$, both are independent:

and X distributioN:

-2 -1 3

0.2 0.3 0.5

and Y distribution:

-1 0 2

0.2 0.4 0.4

I have to create distribution and then I have to calculate expected value and variance of $3X - 2Y$, $X^2 + Y^2$, $X \cdot Y$.

I do not understand how to create distribution of $3X - 2Y$, $X^2 + Y^2$ and $X \cdot Y$ accordingly. After creating distribution, I can calculate Expected value and variance, so I think I do not need any special formulas. However, I cannot understand how $3X - 2Y$ and others should be calculated.

Should I take "each $X$" and "each $Y$" and calculate $3X - 2Y$, so there would be 36 items? I do not really understand how this works and what is a meaning of $3X - 2Y$ if $X$ and $Y$ are two random indpendent variables.

share|improve this question

1 Answer 1

up vote 0 down vote accepted

I'm assuming that you meant $X$ and $Y$ to be independent such that $X$ has probability mass function given by $$ P(X=k)= \begin{cases} 0.5 &\quad \text{if }k=3\\ 0.3 &\quad \text{if }k=-1\\ 0.2 &\quad \text{if }k=-2\\ \end{cases} $$ and $Y$ has probability mass function $$ P(Y=k)= \begin{cases} 0.4 &\quad \text{if }k=2\\ 0.4 &\quad \text{if }k=0\\ 0.2 &\quad \text{if }k=-1\\ \end{cases} $$ Then using the law of the unconscious statistician we obtain $$ E[X]=3\cdot 0.5+(-1)\cdot 0.3+(-2)\cdot 0.2=0.8 $$ and similarly $E[Y]=0.6$. By applying the law again, we can also obtain expressions for $E[X^2]$ and $E[Y^2]$. Having them in hand, we can find the variance of $X$ and $Y$ by the formula $$ \mathrm{Var}(X)=E[X^2]-E[X]^2 $$ and similarly for $Y$.

Now you're able to calculate the mean and variance of $3X-2Y$ by using that $$ E[3X-2Y]=3E[X]-2E[Y], $$ and $$ \mathrm{Var}(3X-2Y)=3^2\mathrm{Var}(X)+(-2)^2\mathrm{Var}(Y) $$ where we explicitly used that $X$ and $Y$ are independent, see e.g. this.

In a similar manner you can compute the mean and variance of $X^2+Y^2$, but to compute the mean and variance of $X\cdot Y$ you need to use that $$ E[g(X)g(Y)]=E[g(X)]\cdot E[g(Y)] $$ for any (measurable) function $g$.

share|improve this answer
    
This makes sense... However, there is also a question to "show distribution for 3X-2Y and others" (Is this some function in form P(3X-2Y = k)? –  renathy Feb 24 '13 at 20:41
    
I mean what makes P(3X-2Y = k)? Every x combination with every y? So total 9 values? –  renathy Feb 25 '13 at 6:01
    
Yes, to find the distribution of e.g. $3X-2Y$ you simply go through all combinations of $(X,Y)$ values. Then you will have to find the probability of $P(3X-2Y=k)$ for $k=-10,-7,-6,-4,-3,-1,5,9,11$. For example$$P(3X-2Y=11)=P(X=3,\,Y=-1)=P(X=3)\cdot P(Y=-1)=0.1,$$ where we have used the independence of $X$ and $Y$. –  Stefan Hansen Feb 25 '13 at 6:20
    
I am sorry, I still need help to calculate E[X^2]... ? –  renathy Feb 25 '13 at 8:06
    
Use the law of the unconscious statistician:$$E[X^2]=3^2\cdot P(X=3)+(-1)^2\cdot P(X=-1)+(-2)^2\cdot P(X=-2)$$ –  Stefan Hansen Feb 25 '13 at 8:07

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.