I have this variable with beta distribution : $Y \sim \mathcal{B}e(\alpha,\frac{1}{3})$.

I have to find the value of $\alpha$ such as : $P(Y \leq 0.416) =0.2 $

Formally for $\alpha \geq 0$ , $\beta \geq 0$ and $0 \leq y \leq 1$ the CDF function of $Y$ at 0.416 is:

$P(Y \leq 0.416) = \frac{\Gamma(\alpha + \beta)}{\Gamma(\alpha)\Gamma(\beta)} \int_0^{0.416} t^{\alpha-1} (1-t)^{\beta-1} dt=0.2$

I am not sure how to proceed. Thanks for the help in advance!!


Maple gives the CDF for $\mathcal{B}e(\alpha,\beta)$ as $$ {\frac {\Gamma \left( \alpha+\beta \right) {y}^{\alpha} {\mbox{$_2$F$_1$}(\alpha,1-\beta;\,1+\alpha;\,y)}}{\Gamma \left( \alpha \right) \Gamma \left( \beta \right) \alpha}} $$ In any case, you need to use numerical methods to solve $F(0.416) = 0.2$. Maple says $\alpha = 0.8563203833$.

  • $\begingroup$ The question is stated as though an analytical solution is expected...but I agree that numerical methods are the way to proceed here. $\endgroup$ – Math1000 Dec 13 '18 at 15:13
  • $\begingroup$ thank you very much for the kind answer. But i don't understand the meaning of this quantity: ${\mbox{$_2$F$_1$}(\alpha,1-\beta;\,1+\alpha;\,y)}$. And how did you obtain the value of $\alpha$? Thank you in advance @Robert Israel $\endgroup$ – andrew Dec 13 '18 at 15:25
  • $\begingroup$ Can i get a solution with the software R? $\endgroup$ – andrew Dec 13 '18 at 16:40
  • $\begingroup$ ${}_2F_1(a,b; c; z)$ is a Gaussian hypergeometric function. The value of $\alpha$ was obtained using a numerical solver (fsolve in Maple). $\endgroup$ – Robert Israel Dec 13 '18 at 18:55

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.