Title: Continuous Random Variables and Their Probability Distributions
1Chapter 4
- Continuous Random Variables and Their Probability
Distributions
2Sec 4.2
3Cumulative Distribution Function
- Definition Let Y be any random variable. The
distribution function (cumulative distribution
function) for Y, F(y), is defined by, F(y)
P(Y?? y). - Note This function is defined for both
continuous and discrete random variables.
4Properties of any cdf F(x)
- Let Y be a random variable with distribution
function F(y). Then - For all a and b, if a lt b then F(a) ? F(b).
(F is a nondecreasing function) - (lower
limit of F is 0) - (upper
limit of F is 1) - (F is
right continuous) - Let Y be be a random variable with distribution
function F(y). Then for a lt b Pa lt Y ? b
F(b) F(a).
5Continuous Random Variables
- Definition A random variable Y is a continuous
random variable if its distribution function F(y)
is continuous for all y?R. - Definition For a continuous random variable Y
with distribution function F(y) the probability
density function (pdf) of Y is defined by f(y)
(d/dy)F(y) F'(y). Consequently,
6Properties of pdf
- If f(y) is the density function of a continuous
random variable Y, then - f(y) ? 0 for all y?R
- ?f(y)dy 1 where the integral is over (-?, ?)
- P(Ya) 0 for any real number a
- If altb, then
7Sec 4.3
8Expected Value of Y
- Definition Let Y be a continuous random
variable with pdf f(y) then the expected value of
Y, E(Y), is defined by E(Y) ? y f(y) dy
provided the integral converges absolutely. - The expected value of Y is denoted by ?.
9Functions of Random Variables
- Theorem Let Y be a continuous random variable.
Suppose that for some real valued function g the
random variable X g(Y) is continuous. Then the
expectation of X is given by E(X) Eg(Y)
? g(y) f (y) dy provided the integral
converges absolutely
10Properties
- Theorem Let c be a constant and let g1(Y),
g2(Y),.., gk(Y) be functions of a continuous
random variable Y. Suppose the expectations of
g1(Y), g2(Y), , gk(Y) all exist. Then, - E(c) c
- Ecg(Y) cEg(Y)
- E(g1(Y) g2(Y) gk(Y) ) E(g1(Y))
E(g2(Y)) E(gk(Y)) - V(Y) E(X - ?)2 E(X2) - ?2.
11Section 4.4-4.7
- Uniform Probability Distribution, U(?1,?2)
- Normal Probability Distribution, N(?,?2)
- Gamma Probability Distribution, Ga(?, ?)
- Chi-Square Probability Distribution, ?2(?)
- Exponential Probability Distribution, exp(?)
- Beta Probability Distribution, Beta(?, ?)
- Uniform Probability Distribution,
U(0,1)Beta(1,1)
12The Uniform Distribution
- The continuous random variable Y has uniform
distribution of the interval (?1, ?2) if the
density function of Y is f(y) 1/(?2 - ?1)
for y ? (?1,?2) and 0 otherwise - E(Y) (?1 ?2)/2
- V(Y) (?2- ?1)2/12
13The Normal Distribution
- The continuous random variable Y has a normal
distribution if the density function of Y is
for all y? (-?,?) - E(Y) ?
- V(Y) ?2
14Calculation of Normal Probability Density
Function (pdf)
- Standard Normal pdf , Denoted by N(0,1)
- Z(Y-?)/?
- Find scores, given probability
- Find proportion/probability, given the scores
15Gamma Distribution
- A random variable Y is said to have a gamma
distribution with parameters ?gt0 and ?0 if and
only if the density function of Y is
16Mean and Variance of Gamma Probability
Distribution
- Mean ?EY??
- Variance ?2VY??2
- MGF my(t)(1-?t)-?
- Special cases
- Chi-square distribution with ? degrees of freedom
if ? ? /2 and ?2 - Exponential distribution when ? 1
17Chi-square Probability Distribution
- A random variable Y is said to have a chi-square
distribution with ? degrees of freedom if and
only if the density function of Y is
Mean ? EY? , Variance ?2VY 2 ? MGF
my(t)(1-2t)-? /2
18Exponential Probability Distribution
- A random variable Y is said to have an
exponential distribution with parameter ?gt0 if
and only if the density function of Y is
Mean ? EY? , Variance ?2 VY ?2 MGF
my(t)(1- ?t)-1 Memoryless property
P(YgtabYgta)P(Ygtb)
19Homework
- p.173-175 4.46, 4.47, 4.54, 4.57, 4.63 p180
4.68, 4.71, 4.72, 4.76
20Beta Probability Distribution
- A random variable Y is said to have a beta
probability distribution with parameters ?gt and
?gt0 if and only if the density function of Y is
21Mean and Variance of Beta Y
- Mean ? EY ?/(??)
- Variance ?2 VY? ? /((??)2(??1))
- MGF no closed form
- Special cases ?? 1, Y is uniform distribution.
- That is f(y)1 , 0?y?1
- 0, elsewhere
22Section 4.9
- Moment
- Moment of Function of Random Variables
23Moment
- Moment If Y is a continuous random variable,
then the kth moment about the origin is given by - ?k EYk, k1,2,
- The kth moment about the mean, or the kth central
moment, is given by - ?k E(Y- ?)k, k1,2,
24Moment Generating Function (MGF)
- MGF If Y is a continuous random variable, then
the moment-generating function of Y is given by - my(t)EetY
- The moment-generating function is said to exist
if there exists a constant bgt0 such that my(t) is
finite for t?b. - The kth moment about origin is
25MGF of Function
- Theorem Let Y be a random variable with density
function f(y) and g(y) be a function of Y. then
the moment-generating function for g(Y) is
26Homework
- p185 4.91, 4.92p193-194 4.104, 4.108, 4.113
27Section 4.10
- Tchebysheffs Theorem
- (Chebyshevs Inequality)
- Let Y be any random variable with finite mean
? and variance ?2. Then, for any kgt0, P(Y- ?
lt k?) ? 1 (1/k2)or P(Y- ? ? k?) ? (1/k2) - Compare it to 68-95-99.7 rules for normal
distribution