[Math] How to find the probability of committing a Type II error

statistics

I'm not sure If I understand what (b) is asking. Does it mean that the alternative hypothesis will be p<0.3, p<0.4, and p<0.5? If it is, then I have to find the probability of committing a type 2 error for each alternatives? If I am right then the setup that I have for p=0.3 in part (b) is correct.

enter image description here

Click here to see the question and my worked out problem

Best Answer

You are on the right track, but you just need to go through and complete the calculation. Note that the sum $$\Pr[X > 3 \mid p = 0.3] = \sum_{x=4}^{10} \binom{10}{x} (0.3)^x (0.7)^{10-x}$$ contains a total of $10-4+1 = 7$ terms, whereas you could save work by exploiting the complementary probability: $$\Pr[X > 3 \mid p = 0.3] = 1 - \Pr[X \le 3 \mid p = 0.3] = 1 - \sum_{x=0}^3 \binom{10}{x} (0.3)^x (0.7)^{10-x},$$ and this latter sum has only $4$ terms, and they are easy to compute: $$\sum_{x=0}^3 \binom{10}{x} (0.3)^x (0.7)^{10-x} = (0.7)^{10} + 10(0.3)(0.7)^9 + 45(0.3)^2(0.7)^8 + 120(0.3)^3(0.7)^7.$$