CAS3 spring 2005 : Q# 23

[piemedia]

Please help me out on this....


Why critical region is X<= C, but not X>= C , because for type I error, the significance level is 2%. Isn't it x>=c for the significance level???

[YuGoHUGO]

the critical region IS X>c, where c=30 accidents... they want to make sure that 98% of the sample falls below this 30 accident mark

[piemedia]

oh,,,,,,, i see,, thank you so much,, good luck on your exam!!!

[carrytheCrøss]

Preface: YugoHugo beat me, but here's the post anyway =0

piemedia, the definition of a Type I Error is P(Reject H0|H0 is true). The null hypothesis in this question is that the proportion of customers who file at least one claim, p, equals 30% (p=30%). Since a discount will be offered if fewer than 30% file at least one claim, the alternative hypothesis is that p is less than 30%.

Now just work from the definition, and note that the proportion X who file at least one claim has mean p and variance p(1-p). And, remember that the

.

The rest, I hope, should follow.