Let X1 and X2 be two independent Binomial variate with parameters (n1,p1) and (n2,p2) respectively. Let Y=X1+X2. Then the MGF (moment generating function) of Y is
"M_Y(t) = M_{X_1}(t)*M_{X_2}(t) = (q_1 +p_1 e^t)^{n_1} * (q_2 +p_2 e^t)^{n_2}"
It is not MGF of Binomial distribution.
But it is MGF of Binomial distribution iif p1 = p2=p:
"(q_1 +p_1 e^t)^{n_1} * (q_2 +p_2 e^t)^{n_2} = (q +p e^t)^{n_1} * (q +p e^t)^{n_2}=(q +p e^t)^{n_1+n_2}"
Now it is MGF of Binomial distribution with parameters (n1+n2, p)
Hence, condition is p1 = p2 then Y=X1+X2∼B(n1+n2, p)
.
Comments
Leave a comment