# Bernoulli Random variable and population problems - i need it for an exam

• Sep 22nd 2012, 12:24 PM
homalina
Bernoulli Random variable and population problems - i need it for an exam
Let Ya and Yb denote Bernoulli random variables from 2 different populations, denoted a and b. suppose that E (Ya)= Pa and E (Yb) = Pb. A random sample of size na is chosen from population a, with sample average denoted P^a and a random sample of size nb is chosen from the population b, with sample average denoted p^b Sample from population a is independent from sample from population b.
Show that E (P^a) = Pa and var (P^a)= Pa(1- Pa)/ na.
Show that E(P^b) = Pb and var (P^b)= Pb(1- Pb)/ nb.

Show that var (P^a - P^b)= Pa(1- Pa)/ na + Pb(1- Pb)/ nb

Suppose that na and nb are large. Show that a 95% confidence interval for Pa - Pb is given by (P^a - P^b) plus and minus 1.96 sqrt P^a(1- P^a)/ na + P^b(1- P^b)/ nb.

How would you construct a 90% confidence interval for Pa - Pb?

I have missed the class when prof. taught these.... :(
• Sep 22nd 2012, 04:57 PM
harish21
Re: Bernoulli Random variable and population problems - i need it for an exam
Quote:

Originally Posted by homalina
Let Ya and Yb denote Bernoulli random variables from 2 different populations, denoted a and b. suppose that E (Ya)= Pa and E (Yb) = Pb. A random sample of size na is chosen from population a, with sample average denoted P^a and a random sample of size nb is chosen from the population b, with sample average denoted p^b Sample from population a is independent from sample from population b.
Show that E (P^a) = Pa and var (P^a)= Pa(1- Pa)/ na.
Show that E(P^b) = Pb and var (P^b)= Pb(1- Pb)/ nb.

Show that var (P^a - P^b)= Pa(1- Pa)/ na + Pb(1- Pb)/ nb

Suppose that na and nb are large. Show that a 95% confidence interval for Pa - Pb is given by (P^a - P^b) plus and minus 1.96 sqrt P^a(1- P^a)/ na + P^b(1- P^b)/ nb.

How would you construct a 90% confidence interval for Pa - Pb?

I have missed the class when prof. taught these.... :(

I have given you hints on finding the expected value and variance of sample average in one of your previous problems..use the same concept here