PBG 650 Advanced Plant Breeding Mathematical Statistics Concepts Probability laws Binomial distributions Mean and Variance of Linear Functions
Probability laws the probability that either A or B occurs (union) the probability that both A and B occur (intersection) the joint probability of A and B the conditional probability of B given A
Statistical Independence If events A and B are independent, then
Bayes’ Theorum (Bayes’ Rule) Conditional probability Bayes’ Theorum . Pr(A) is called the prior probability Pr(A|B) is called the posterior probability
Probabilities Plant Height A1A1 A1A2 A2A2 Marginal Prob. height ≤ 50 cm 0.10 0.14 0.06 0.30 50 < height ≤ 75 0.04 0.18 0.32 height > 75 cm 0.02 0.16 0.20 0.38 0.48 0.36 1.00 Marginal Probability: Pr(Genotype= A1A1) = 0.16 Joint Probability: Pr(Genotype= A1A1, height ≤ 50) = 0.10 Conditional Probability:
Statistical Independence If X is statistically independent of Y, then their joint probability is equal to the product of the marginal probabilities of X and Y Plant Height A1A1 A1A2 A2A2 Marginal Prob. height ≤ 50 cm 0.0480 0.1440 0.1080 0.30 50 < height ≤ 75 0.0512 0.1536 0.1152 0.32 height > 75 cm 0.0608 0.1824 0.1368 0.38 0.16 0.48 0.36 1.00
Discrete probability distributions Let x be a discrete random variable that can take on a value Xi, where i = 1, 2, 3,… The probability distribution of x is described by specifying Pi = Pr(Xi) for every possible value of Xi 0 ≤ Pr(Xi) ≤ 1 for all values of Xi ΣiPi = 1 The expected value of X is E(X) = ΣiXiPr(Xi) =X
Binomial Probability Function A Bernoulli random variable can have a value of one or zero. The Pr(X=1) = p, which can be viewed as the probability of success. The Pr(X=0) is 1-p. A binomial distribution is derived from a series of independent Bernouli trials. Let n be the number of trials and y be the number of successes. Calculate the number of ways to obtain that result: Calculate the probability of that result: Probability Function
Binomial Distribution Average = np = 20*0.5 = 10 Variance = np(1-p) = 20*0.5*(1-0.5) = 5 For a normal distribution, the variance is independent of the mean For a binomial distribution, the variance changes with the mean
Mean and variance of linear functions Mean and variance of a constant (c) Adding a constant (c) to a random variable Xi the mean increases by the value of the constant the variance remains the same
Mean and variance of linear functions Multiplying a random variable by a constant Adding two random variables X and Y multiply the mean by the constant multiply the variance by the square of the constant mean of the sum is the sum of the means variance of the sum the sum of the variances if the variables are independent
Variance - definition The variance of variable X Usual formula Formula for frequency data (weighted)
Covariance - definition The covariance of variable X and variable Y Usual formula Formula for frequency data (weighted)