2
$\begingroup$

I have 2 questions where I have to use the Pareto distribution.

  1. A family of pdf’s that has been used to approximate the distribution of income, city population size, and size of firms is the Pareto family. The family has two parameters, $k$ and $y$, both $> 0$, and the pdf is: $$ f(x;k,\theta)=\frac{k\cdot\theta^k}{x^{k+1}} \ for \ x\geq\theta $$ and is $ 0 $ otherwise.

a) Sketch the graph of $ f(x;k,\theta) $

b) Verify the the total area under the graph is $ 1 $.

c) For $ \theta<a<b $, obtain an expression for the probability $P(a\leq{X}\leq{b}). $

For this question, I have integrated and proved part b) because all $k $ and $\theta$ cancel. I have also got an expression for part c) by just changing the bounds to $ a $ and $b $. However, I have no idea how to graph the 3 variables, any thoughts?

  1. Let X have the Pareto pdf introduced in Exercise 1.

a) If $k>1$, compute $E(X)$

b) What can you say about $E(X)$ if $k=1?$

c) If $k>2$, show that $ V(X)=k\theta^2(k-1)^{-2}(k-2)^{-1} $.

d) If $k=2$, what can you say about $V(C)$?

e) What conditions on $k$ are necessary to make $ E(X^n) $finite?

I got $E(X)=\int_\theta^\infty{x\frac{k\theta^k}{x^{k+1}}}dx=k\theta^k\int_\theta^\infty{x^{-k}}dx=-\frac{k\theta}{-k+1} $. And I got $E(X^2)=\int_\theta^\infty{x^2\frac{k\theta^k}{x^{k+1}}}dx=k\theta^k\int_\theta^\infty{x^{-k+1}dx}=\frac{-k\theta^2}{-k+2}$

$\endgroup$
  • $\begingroup$ Please see the wiki page of the Pareto distribution which likely has all the answers to your questions. $\endgroup$ – StubbornAtom Nov 9 '18 at 17:53
  • $\begingroup$ Right, I got the graph, but now my E(X) and V(X) for question 2 aren't matching what they give V(X) to be, any help? $\endgroup$ – D. Wei Nov 9 '18 at 18:23
  • $\begingroup$ Yes, but you have to show your work. $\endgroup$ – StubbornAtom Nov 9 '18 at 18:24
  • $\begingroup$ By the way, your answers are correct now if you mention the conditions under which the moments exist. $\endgroup$ – StubbornAtom Nov 9 '18 at 18:59
1
$\begingroup$

In light of the fact that $$\frac{1}{x^\alpha}\to 0\text{ as }x\to\infty\quad\text{ provided } \alpha>0$$

, we have the $r$-th order raw moment of $X$ about $0$ :

\begin{align} E(X^r)&=\int_{\theta}^\infty \frac{x^r\,k\theta^k}{x^{k+1}}\,dx \\&=k\theta^k\int_{\theta}^\infty x^{r-k-1}\,dx \\&=k\theta^k\lim_{A\to\infty}\left[\frac{x^{-(k-r)}}{-(k-r)}\right]_{\theta}^A \\&=\frac{k\theta^r}{k-r}\qquad,\text{ if }k>r \end{align}

You should be able to proceed now.

$\endgroup$
  • $\begingroup$ Right so I get $ V(X)=\frac{k\theta^2}{k-2}-(\frac{k\theta}{k-1})^2=k\theta^2(\frac{1}{k-2}-\frac{k}{(k-1)^2})=k\theta^2(\frac{k^2-2k+1-k^2-2k}{(k-2)(k-1)^2})$. Thanks! $\endgroup$ – D. Wei Nov 10 '18 at 3:12
  • $\begingroup$ @D.Wei The second $2k$ in the numerator has a $+$ in front, so variance is just $\frac{k\theta^2}{(k-2)(k-1)^2}$ for $k>2$. $\endgroup$ – StubbornAtom Nov 10 '18 at 6:12
  • $\begingroup$ Yes, thank you for pointing that out, that was just a typo though, thanks to your help I have solved this problem :D $\endgroup$ – D. Wei Nov 12 '18 at 16:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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