Matlab在金融工程中的应用(peng liu)Lecture3_Matlab.pdfVIP

Matlab Workshop MFE 2006 Lecture 3 Stefano Corradin Peng Liu /peliu/computing Haas School of Business, Berkeley, MFE 2006 Applications in Finance I 3.1 Monte Carlo simulation of known results. 3.2 Monte Carlo simulation: generating sample paths. 3.3 Monte Carlo simulation: option pricing. 3.4 Monte Carlo simulation: puzzles. Haas School of Business, Berkeley, MFE 2006 1 Monte Carlo simulation of known results The idea of Monte Carlo simulation is to generate samples from a distribution by using a computer. This technique is very useful and frequently used in finance when there is not a closed expression. Central limit theorem: we generate 1, 000 samples of size 50 from an exponential distribution with mean 1/λ = 2 E = exprnd(2,50,1000); every column represents a different sample. Let calculate the mean and plot a histogram of the mean values m = mean(E); hist(m) Haas School of Business, Berkeley, MFE 2006 2 We know that the mean is µ = 1/λ = 2 and the variance is σ2 = 1/λ2 = 4. Haas School of Business, Berkeley, MFE 2006 3 What is the mean and variance of the sample mean m? mean(m) ans = 1.9945 var(m) ans = 0.0858 What is the sample mean of the variance? v=var(E); mean(v) ans = 4.0347 Haas School of Business, Berkeley, MFE 2006 4 Properties of a random sample from a normal distribution: Given a ran- dom sample X , X , . . . , X from a normal distribution N (µ, σ) and let 1 2 n ¯ 1 n