Transcript Binomial
The Binomial Distribution Permutations: How many different pairs of two items are possible from these four letters: L, M. N, P. L,M L,N L,P M,L M,N M,P N,L N,M N,P P,L P,M P,N Pr N N! ( N r )! 4! P (4 2)! 4 2 4 X 3(2!) P (2)! 4 2 P24 12 P = # of Permutations N = # of items r = how many taken at a time Combinations L,M L,N L,P M,N M,P N,P (order is not important) CrN CrN N! r !( N r )! C = # of Combinations N = # of items r = how many taken at a time 4! 2!(4 2)! 4 X 3 X 2 X 1 24 C 6 2 X 1X 2 X 1 4 N r If we flip a single coin, there are two possible outcomes: head or tail (H or T). P(one head) = .5 P(no head) = .5 Together the probabilities of these possible outcomes sum to 1.0. If we flip two coins there are four possible outcomes. H,H H,T T,H T,T P(two heads) = .25 P(one head) = .50 P(no heads) = .25 Total = 1.0 Can be calculated by… P = P(H) and Q = P(T) P2 ( P Q) 2 ( P Q)( P Q) Or P 2 PQ Q 2 2PQ Q2 Probability of 2 heads Probability of 1head and 1 tail Probability of 2 tails 2 * Think about the frequency distribution. Flip 3 Coins: 8 possible outcomes H,H,H H,H,T H,T,H T,H,H T,T,H T,H,T H,T,T T,T,T P(3 heads) = 1/8 or .13 P(2 heads) = 3/8 or .37 P(1 head) = 3/8 or .37 P(0 heads) = 1/8 or .37 Total = 1.0 Can be calculated by….. ( P Q) 3 or ( P Q)( P Q)( P Q) P 3 3P 2 Q 3PQ2 Q3 Binomial Distribution A binomial distribution is produced when each of a number of Independent trials results in one of two Mutually Exclusive outcomes. A single flip of a coin is a Bernoulli trial. They reflect a discrete variable. P( X ) CrN P X Q( N X ) N! p X Q( N X ) X !( N X )! P(X) = Probability of X number of an outcome N = Number of trials P = Probability of success on any given trial Q = (1 – P) CXN The number of combinations of N things taken X at a time Example: Guess if I am thinking of the colour black or white. Because I randomly choose one of the two colours, the probability that I am thinking of black is .5 on any given trial. There will be 10 trials. What is the probability of guessing 8 of them correctly? 10! P(8) (.58 )(.52 ) 8!(2!) 10 * 9(8!) (.0039062)(.25) 8!(2!) = 45(.009765) = .0439425 This is the probability of * OR MORE correct, but of exactly 8 out of 10 correct. Binomial Distribution & Testing a Hypothesis Let us repeat the previous example, but now the question is what is the probability of guessing 8 or more of the trials correctly. We know that P(8) = .44 That is less than .05, But that is P(8) against 10, including P(9) and P(10). We need to test the probability of 8 or more correct. P(8) = .044 P(9) = .010 P(10) = .001 Total P = .055 The probability of guessing 8 or more correctly out of 10 turns out to have a probability greater than .05, thus we fail to reject the null hypothesis. Which was what? Binomial Distribution We can calculate the probability of 0 to 10 correct. # Correct Probability 0 1 2 3 4 5 6 7 8 9 10 .001 .010 .044 .117 .205 .246 .205 .117 .044 .010 .001 0.25 0.2 0.15 P 0.1 0.05 0 0 1 2 3 4 5 6 7 8 9 10 Mathematical Sampling Distribution As N gets large, all binomial distributions approach normality, regardless of P. Thus, Mean = NP For example, 10(.5) = 5 That is, if you examine the data on the previous slide, you will see that the mean of the distribution is 5 (the number of correct trials). Variance = NPQ Standard Deviation = For example, 10(.5)(.05) = 2.25 NPQ = 1.58