Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Math/Stat 360-1: Probability and Statistics, Washington State University Haijun Li lih@math.wsu.edu Department of Mathematics Washington State University Week 5 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 1 / 15 Outline 1 Section 4-1: Probability Density Functions 2 Section 4-2: Cumulative Distribution Functions and Expected Values Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 2 / 15 Distribution of Continuous Random Variables Random Variables (RV): Denoted by capital letters, X , Y , N, ... Values of Random Variables: Denoted by small letters, x, y , n, ... Distribution of a continuous RV X : Relative frequencies of various values of X . Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 3 / 15 S&P 500 Monthly Returns (10/1965 - 7/2005) S&P 500 Monthly Return = random variable X. http://www.optionetics.com/images/articles The distribution = continuous version of the histogram of X . 9-6-06 2.gif (GIF Image, 572 × 419 pixels) Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 4 / 15 Probability Density Function Probability Density Function (PDF) f (x) describes the probability density of a continuous RV X at x. 1 f (x) ≥ 0 R∞ 2 f (x)dx = 1PROBABILITY DENSITY FUNCTIONS RIBUTIONS AND −∞ Rb 3 P(a < X < b) = a f (x)dx = the area under f (x) between a and b. f (x) P(a < X < b) a Haijun Li b Math/Stat 360-1: Probability and Statistics, Washington State University x Week 5 5 / 15 Probability Density Function PDF describes relative frequency of a continuous random variable X . P(X = a) = 0. P(a ≤ X ≤ b) = P(a < X ≤ b) = P(a ≤ X < b) = P(a < X < b). rb12_6.gif (GIF Image, 613 × 474 pixels) Haijun Li http://www.weibull.com/hotwire/issue12/rb12_6.gif Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 6 / 15 Example The PDF of the failure time X of an electronic component in a copier (in hours) is h 1 x i f (x) = exp − , x ≥ 0, 3000 3000 and zero otherwise. Find the probability that 1 2 3 A component lasts than 1000 hours before failure. R ∞ more x x 1 P(X > 1000) = 1000 3000 e− 3000 dx = [−e− 3000 ]∞ 1000 = 0.716. A component fails between 1000 and 2000 hours. R 2000 1 − x x P(1000 < X < 2000) = 1000 3000 e 3000 dx = [−e− 3000 ]2000 1000 = 0.203. Determine the number of hours at which 10% of all components have failed. a Find a such that P(X ≤ a) = 0.1. That is, 1 − e− 3000 = 0.1. Solve this, we have a ≈ 316 hours. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 7 / 15 Cumulative Distribution Function = Cumulative Frequency Cumulative Distribution Function (CDF) of X Z x F (x) = P(X ≤ x) = f (u)du. −∞ 1 2 3 F (x) is non-decreasing. 0 ≤ F (x) ≤ 1. F (x) → 1 as x → +∞. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 8 / 15 Figure : Comparison of PDF and CDF Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 9 / 15 Rb 1. P(a ≤ X ≤ b) = F (b) −http://http-server.carleton.ca/~gkardos/88403/Reliability/RE F (a) = a f (x)dx 2. f (x) = F 0(x) 2.gif (GIF Image, 600 × 427 pixels) Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 10 / 15 Example Consider a PDF f (x) = h 1 x i exp − , x ≥ 0, 3000 3000 and zero otherwise. Find the CDF. Z x u x 1 F (x) = e− 3000 du = 1 − e− 3000 , x ≥ 0. 0 3000 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 11 / 15 Example Consider the PDF of the uniform RV over [a, b]. f (x) = 1 , a ≤ x ≤ b, b−a and zero otherwise. Find the CDF. Rx 1 F (x) = a b−a du = x−a , a ≤ x ≤ b. b−a F (x) = 0, x < a. F (x) = 1, b < x. Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 12 / 15 Mean and Variance Mean of a continuous RV X Z ∞ Z E(X ) = µ = xf (x)dx = −∞ ∞ x −∞ Variance of a continuous RV X Z ∞ Z 2 2 V (X ) = σ = (x−µ) f (x)dx = −∞ Standard deviation (SD) of X : σ = R∞ σ 2 = −∞ x 2 f (x)dx − µ2 . Haijun Li f (x)dx . | {z } prob. mass at x ∞ (x−µ)2 −∞ p f (x)dx | {z } prob. mass at x V (X ). Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 13 / 15 Figure : σy = SD of Y . Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 14 / 15 Example Suppose that the CDF of the length (in millimeters) of computer cables is x ≤ 1200 0 0.1x − 120 1200 < x ≤ 1210 F (x) = 1 x > 1210 1 Find P(X < 1208), and P(1195 < X < 1205). P(X < 1208) = F (1208) = 0.8 P(1195 < X < 1205) = F (1205)−F (1195) = F (1205) = 0.5. 2 What is its PDF and mean? f (x) = F 0 (x) = 0.1, 1200 < x ≤ 1210. Z 1210 E(X ) = 0.1xdx = [0.05x 2 ]1210 1200 = 1205. 1200 Haijun Li Math/Stat 360-1: Probability and Statistics, Washington State University Week 5 15 / 15