1)-Any one tell me ,how to find the value P- value of 1.21= F(11,11) distribution in the table. 2)-Any oneution of following question is correct or not : Question: A box contains 8 bulbs of which 3 are defective .A bulb is selected from the box and tested if it is defective another bulbs is selscted and tested untill a non-defective bulb is chosen. Find the expected number of bulbs chosen? E[x]= sumation [x*p*q] , x>0 (geometric type 2 distribution) =pq+2pq^2 +3pq^3 +.............. =( pq)/(1-q)^2 =q/p where p=3/8 =5/8 =1.666
In the current (gold) tables F11,11 is not listed. However we could interpolate between F10,11 and F12,11. However, since the P(F11,11>2.228) =0.1 we can see that P(F11,F11>1.21) > 0.1 Since we don't have percentage points tables higher than 10% we can't look this value up. However a P-value of 10% (or 20% if it is a two-sided test) both would give a conclusion of "do not reject H0". A couple of things are not clear on this question. Are the bulbs being replaced or not? If they are then you are right it is a geometric (which has a fixed probability of success). If they are not then you have a hypergeometric distribution. Secondly are you counting the non-defective bulb in the expected number of bulbs chosen? If so (and we're replacing the bulbs) - then you are counting trials which is a type 1 geometric. If not (and we're replacing the bulbs) - the you are counting failures which is a type 2 geometric.