1
$\begingroup$

This problem was part of an assignment in my undergrad stats class.
Suppose Steph Curry takes ten 3-point shots per game, and the number of 3-point shots scored follows a $𝐵𝐼𝑁\sim(10, 𝑝)$ distribution, where 𝑝 is the probability of successfully scoring a 3-point shot. [Note: in basketball a 3-point shot is one taken from behind the 3- point arc]. To estimate this probability, a random sample of 10 games is considered and the number of successful 3-point shots is recorded. These data are shown below.

$$\begin{array}{ccccc} \hline 5 & 3 & 0 & 7 & 10 \\ 2 & 4 & 6 & 8 & 1 \\ \hline \end{array}$$

Calculate the maximum likelihood estimate, $\tilde{p}_{ML}$, of $p$, the probability of scoring a 3- point shot, and verify (using either the first or second derivative test) that you have found a maximum.

So I've $X\sim BIN(10,p)$

I go through the steps to find the estimator and arrive at

\begin{equation} \sum_{i=1}^{10}x_i(1-p)=p\sum_{i=1}^{10}(10-x_i) \end{equation}

I'm at a standstill in how I solve for $p$. Then when I have the estimator, do I do anything with the provided info in the table? any help would be appreciated

$\endgroup$
1
$\begingroup$
  1. Start by bringing the $1-p$ term outside the sum, $$ (1-p)\sum_{i=1}^{10} x_i = p\sum_{i=1}^{10}(10-x_i)$$

Upon using the distributive property on $(1-p)\sum_{i=1}^{10}x_i$, you will see that both sides of the equation have a $-p\sum_{i=1}^{10}x_i$ that can cancel out. You should be able to solve for $p$ from there. You may also need to know that $\sum_{i=1}^{a}b = a*b$, where $a$ is a fixed positive integer and $b$ is a fixed arbitrary real number.

  1. Don't forget to apply the second derivative test to show that the critical point is a maximum.

3.To compute the MLE estimate: When you solve for $p$, you will get a formula for $\tilde{p}_{ML}$. The numbers in the table correspond to the values of $x_i$ in that table. Plug them in.

Hope this helps.

$\endgroup$
3
  • $\begingroup$ Thank you so much for your help, I ended up getting $\tilde{p}=\dfrac{\sum_{i=1}^{10}x_i}{10}$. Does this seem right? When I plug in the $x_i$'s I get 4.6. Is this the amount of 3's we expect him to make per game? $\endgroup$
    – Lanous
    Feb 27 '17 at 18:23
  • 1
    $\begingroup$ Close but no cigar. $p$ is a probability, it should be between 0 and 1. Hence, we would hope $\tilde{p}$ is also between 0 and 1. You are missing a multiplicative factor. Double check your algebra. $\endgroup$
    – RMurphy
    Feb 27 '17 at 20:34
  • 1
    $\begingroup$ when I distribute the sum on the right hand side does the $10$ become $10\cdot 10$ which would give me $100$ in the denominator. I think you point this out above. But I overlooked it until now. thank you $\endgroup$
    – Lanous
    Feb 28 '17 at 4:18

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.