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.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am doing self-study. but I don`t calculate this problem. so I want to calculating course.

Three numbers are randomly selected and rounded-off to the nearest interger. Let $X_1, X_2, X_3$ denote the round-off errors and assume that they are independent and uniformly distributed over $[-0.4,0.4]$. Let $D$ be the sum of the errors: $D=X_1+X_2+X_3$

1) Compute $E(D)$

2) Compute $\mathrm{Var}(D)$

3) Use the Central Limit Theorem to approximate the probability $P(D>0.1)$. Give the answer in terms of the cdf $\Phi(z)=P(Z<-z)$ of $Z\sim N(0,1)$.

share|cite|improve this question
up vote 1 down vote accepted

We first calculate the mean and variance of $X_1$. By symmetry, the mean is $0$. If roundoff errors are indeed uniformly distributed on $[-0.4,0.4]$, then the density function of $X_1$ is $\dfrac{1}{0.8}$ over this interval, and $0$ elsewhere. The variance of $X_1$ is $E(X_1^2)-(E(X_1))^2$. We have $$E(X_1^2)=\int_{-0.4}^{0.4}\frac{1}{0.8}x^2\,dx.$$ Integrate. We get $\dfrac{(0.4)^2}{3}$.

Thus $X_1$ has variance $\dfrac{(0.4)^2}{3}$. So do $X_2$ and $X_3$. (You may have been given a formula for the variance of a uniform on $[a,b]$. In that case, you could just use that formula.)

Since $D=X_1+X_2+X_3$, we see that $D$ has mean $0$ and variance $(0.4)^2$, and therefore standard deviation $0.4$.

Now you are asked to use CLT to approximate the probability that $D\gt 0.1$. This is a straighforward normal distribution mean $0$ standard deviation $0.4$ calculation. I am a bit uncomfortable with using CLT, since $3$ is a very small sample size. The answer, if we use CLT, turns out to be $1-\Pr\left(Z\le \dfrac{0.1}{0.4}\right)$.

share|cite|improve this answer
Thank you so much Could you answer my other question? – user72446 Dec 10 '12 at 7:36
It is getting very late here! I am hoping that by the time I wake up, someone else will have done it. Your other questions were more of a statistical nature. For such questions, you may get better answers if you ask on the stats Stack exchange (can find a link at bottom of page). The questions were standard, and the stats site probably already has answers to very similar questions. – André Nicolas Dec 10 '12 at 7:52

Your Answer


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.