This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 2 EXPONENTIAL AND GAMMA DISTRIBUTIONS Lecture 15 ORIE3500/5500 Summer2011 Li 1 Multivariate Normal Distribution The bivariate normal distribution is one of the most important joint distri butions. ( X,Y ) are said to be a bivariate normal vector is they have density, for all ,∞ < x,y < ∞ , f X,Y ( x,y ) = 1 2 πσ 1 σ 2 p 1 ρ 2 exp ‡ ( x μ 1 ) 2 σ 2 1 2 ρ ( x μ 1 )( y μ 2 ) σ 1 σ 2 + ( y μ 2 ) 2 σ 2 2 2(1 ρ 2 ) · . The parameters μ 1 ,μ 2 ,σ 1 ,σ 2 and ρ , characterize the distribution. μ 1 and μ 2 are the means of X and Y respectively and σ 2 1 , σ 2 2 are the variance of X and Y respectively. ρ is the correlation between X and Y . We know that in general if two variables are independent then they have 0 correlation. But the other way is not true. That is , if two random variables are uncorrelated then they are not necessarily independent. But if ( X,Y ) have bivariate joint normal distribution and they are uncorrelated, then they are independent. This is a property of the normal distribution. The multivariate normal distribution is characterized by its mean and variancecovariance matrix. 2 Exponential and Gamma distributions Although the normal distribution finds widespread application in many areas, it can not be used to model everything. For example, the waiting time of a person in a queue, the service time of a customer in a counter, duration of a phone call etc. can not be modeled by a normal random variable for the simple fact that these are all nonnegative random variables. In some such cases, the exponential distribution or the gamma distribution can be used. A continuous random variable is said to follow exponential distribution with parameter λ , or X ∼ exp( λ ), if it has density f X ( x ) = λe λx ,x ≥ . 1 2 EXPONENTIAL AND GAMMA DISTRIBUTIONS The cdf of X is F X ( x ) = 1 e λx ,x ≥ . A random variable having an exp(1) distribution is said to be a standard exponential random variable. We compute the moment generating function of X ∼ exp( λ ) φ X ( t ) = Z ∞ e tx λe λx dx = ‰ λ λ t t < λ ∞ otherwise....
View
Full
Document
This note was uploaded on 12/09/2011 for the course IMSE 0301 taught by Professor Song during the Spring '11 term at HKU.
 Spring '11
 SONG

Click to edit the document details