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 have a random variable $T$ defined as $T=\frac{1}{n}(X_1+\cdots+X_n)$ with $X_i$ being $N(\mu,1)$ distributed. I want to know how $T$ is distributed. This is what I did but I am not sure whether its good: I know that the sum of $n$ normal distributed random variables is distributed $N(n\mu,n\sigma^2)$. So $X_1+\cdots+X_n=n\mu+n\sigma U$ with $U$ being the standard normal distribution. So $T=\mu+U$.

Edit:I think something is wrong because $T$ is indepent of $n$. So what is wrong?

Is this the good way to solve such questions?

share|improve this question
1  
$X_1,\ldots,X_n$ are independent, right? –  Stefan Hansen Jan 20 '13 at 15:53
    
@StefanHansen yes they are independent. –  Badshah Jan 20 '13 at 15:54

1 Answer 1

up vote 1 down vote accepted

This is not entirely true. You have that $X_1+\cdots +X_n=n\mu +\sqrt{n}\sigma U$ for some random variable $U$ following a standard normal distribution, and so $T=\mu+\frac{\sigma}{\sqrt{n}}U$.

However, I like doing it this way where you don't have to introduce $U$: $$ X_1+\cdots X_n\sim N(n\mu,n\sigma^2), $$ where $\sim$ means "is distributed" and so $$ T=\frac{1}{n}\left(X_1+\cdots+X_n\right)\sim N\left(\mu,\frac{1}{n}\sigma^2\right). $$ This way you're only talking about the distribution of $T$ and not about any pointwise equality between $T$ and some other random variable $U$.

Another way to go about it is to note that, by independence, we have that $(X_1,\ldots,X_n)$ follows a multivarite normal distribution and so we know that any linear transformation of this vector is again a normal distribution. And since $T=f(X_1,\ldots,X_n)$, where $f$ is the linear function $$f(x_1,\ldots,x_n)=\frac{1}{n}(x_1+\cdots+x_n),$$ we already know that $T$ is normal distributed. So we just have to determine its mean and variance, and this can be done by properties of the expectation, i.e. $$ E[T]=\frac{1}{n}\left(E[X_1]+\cdots +E[X_n]\right)=\frac{1}{n}nE[X_1]=\mu, $$ and by independence $$ \mathrm{Var}(T)=\frac{1}{n^2}\left(\mathrm{Var}(X_1)+\cdots+\mathrm{Var}(X_n)\right)=\frac{1}{n^2}n\sigma^2=\frac{1}{n}\sigma^2. $$ Hence $T\sim N(\mu,\frac{1}{n}\sigma^2)$.

share|improve this answer

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.