2
$\begingroup$

I was reading up on finding the cdf from a pdf, and noticed that my cdf did not satisfy the condition of $F_Y{(−1)}=0$ and $F_Y{(1)} = 1$. Given the equation $f_Y(y) = cy^2(1-y)1_{[0,1]}(y)$ I calculated the cdf to be $\frac{1}{12}(4-3x)x^3$ by integrating the pdf: $\int_{0}^{x}\frac{1}{12}y^2(1-y)dy$. However this integral does not satisfy the above conditions, so I was hoping to see if someone could point out what I did wrong.

$\endgroup$
  • $\begingroup$ "$F_Y{(−1)}=0$" should be $F_Y{(0)}=0$ because the support of the pdf is $[0,1]$. $\endgroup$ – Jean Marie Oct 8 '17 at 20:06
  • $\begingroup$ @JeanMarie Ah i did not know that. So the cdf should be 0 for the endpoints of the range? $\endgroup$ – Curious Student Oct 8 '17 at 20:08
  • 1
    $\begingroup$ only for the leftmost one. for the rightmost one it should be 1. $\endgroup$ – mathreadler Oct 8 '17 at 20:10
  • $\begingroup$ @jean-marie: Then there would be no need for $1_{[0,1]}$ $\endgroup$ – gammatester Oct 8 '17 at 20:12
  • $\begingroup$ @gammatester No : $1_{[0,1]}$ should still be necessary for the pdf. $\endgroup$ – Jean Marie Oct 8 '17 at 20:14
1
$\begingroup$

The pdf is $12 \cdot y^2 (1-y)$. You have the wrong proportionality constant.

$\endgroup$
  • $\begingroup$ I am not a smart man. So is the correct cdf $(4-3x)x^3$? $\endgroup$ – Curious Student Oct 8 '17 at 20:07
  • 1
    $\begingroup$ yes, that is correct. $\endgroup$ – Srikant Oct 8 '17 at 20:08
  • $\begingroup$ Looks like the support is $[-1,1],$ then the PDF is is $f_Y(y) = 12y^2(1-y)1_{[0,1]}(y).$ Your function has $\int_{-1}^1 = 8$ $\endgroup$ – gammatester Oct 8 '17 at 20:10
  • $\begingroup$ @gammatester I think the support is [0, 1] as indicated by the indicator function in $f(y)$. $\endgroup$ – Srikant Oct 8 '17 at 20:12

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.