Lec3 3.

Spring, 2005

Random variables
Denition of Random Variable Discrete Random Variables Continuous Random Variables Probability Distributions Events of Probability Zero The Mean of a Random Variable Denition Rules for the Mean The Variance of a Random Variable Denition Rules for the Mean Covariance and Correlation
1

 Joint, Marginal and Conditional Distributions The Binomial Distribution

Denition of Random Variable A random variable is a variable whose value is a numerical outcome of a random phenomenon. Technically, it is actually a function that maps elements of the sample space S into numerical outcomes. Example: Tossing a coin four times HHHH HHHT HHT H HHT T S= . . . . . . TTTH TTTT Let X denote the number of heads. For example, given the outcome is HTTH, X = X ( ) = 2
3

Given the outcome is HTTT, X = X ( ) = 1 Then X is a random variable.

We usually denote the random variables by capital letters near the end of the alphabet, such as X or Y , and denote the realized values by the corresponding lower case letter, such as x or y . There are two types of random variables: A discrete random variable takes values in a nite or countable set, for example, in {1, 2, 3, 4, 5} or all integer values. A continuous random variable takes values in a continuum, for example, in an interval [0, 1].

A Discrete Random Variable Consider rolling two dice. (1, 1) (1, 6) . . ... . . S= . . (6, 1) (6, 6) Let X be the total value of the roll. Thus, X ((a, b)) = a + b, (a, b) S X , then, is a discrete random variable taking values 1, 2, . . . , 12.

A Continuous Random Variable Consider measuring the diameter of ball bearings coming o an assembly line. S = possible ball bearings Given each possible ball bearing , let Y ( ) be the diameter of . Clearly, Y > 0, but any positive diameter is theoretically possible. That is, Y takes values in (0, ). Thus, Y is a continuous random variable.

Probability Distributions The probability distribution of a random variable X tells us what values X can take and how to assign probabilities to those values. Discrete random variables The probability distribution of a discrete random variable X lists the possible values of X and their probabilities. X Probability x1 p1 x2 p2 x3 p3 ... ... xk pk

The probabilities pi must satisfy: 1. 0 pi 1 for all i. 2. p1 + p2 + + pk = 1.

Discrete Random Variables Discrete Uniform example : Let X be the result of a fair dice. The probability distribution of X is: X Probability 1
1 6

2
1 6

3
1 6

4
1 6

5
1 6

6
1 6

Bernoulli example : Bernoulli trial is an experiment with only two possible outcomes, success or failure. Let X = 1 indicate success with probability p and X = 0 indicate failure with probability 1 p. Then X follows Bernoulli distribution with parameter p.

For example, let X be the number of head of a fair toss, then 1 X Bernoulli( 2 ). The probability distribution of X is: X Probability 0
1 2

1
1 2

10

Example: Two Rolls of a Die (cont.) Recall X denotes the total value of the two rolls. 0 6 P (X = 1) = P (X = 7) = 36 36 1 5 P (X = 2) = P (X = 8) = 36 36 2 4 P (X = 3) = P (X = 9) = 36 36 3 3 P (X = 4) = P (X = 10) = 36 36 4 2 P (X = 5) = P (X = 11) = 36 36 5 1 P (X = 6) = P (X = 12) = 36 36 1 2 3 1 P (X 3) = + = = 36 36 36 12 6 5 4 3 2 1 P (7 X 11) = + + + + = 36 36 36 36 36 12
11

Continuous Random Variables The probability distribution of a continuous random variable X is described by a density curve f (x). The probability of any event is the area under the density curve f (x) and above the values of X that make up the event. The density curve f (x) must satisfy: 1. f (x) 0, for all x. 2. The total area under the curve is 1. Example Suppose the lifetime of light bulbs X follows a normal distribution with mean 1,000 hours and standard deviation 150 hours.
12

 P (X > 1000) = ? P (X 750) =? P (X = 1000) = ?

13

Events of Probability Zero Note that a continuous probability distribution assigns probability 0 to any individual value. This is clear if we observe that for any function f and any value x,
x

f (x)dx = 0
x

For this reason, when working with continuous probability distributions only, P (X x) = P (X < x) P (X x) = P (X > x)

14

The Mean of a Random Variable Suppose X is a discrete random variable with k possible values, x1 , x2 , . . . , xk . Let pi = P (X = xi ). The mean or expected value of X is given by X = E (X ) = p1 x1 + p2 x2 + + pk xk
k

=
i=1

pi xi

15

If Y is a continuous random variable with density function f (y ), then the mean is given by Y = E (Y )

yf (y ) dy

(The book sweeps this under the rug, calling it advanced mathematics.)

16

Rules for Means Let X and Y be random variables, and let a, b and c be xed numbers, then E (aX + bY + c) = aE (X ) + bE (Y ) + c aX +bY +c = aX + bY + c That is, expectation is a linear operator. Furthermore, for any constant c, E (c) = c.

17

Variance of a Random Variable Suppose X is a discrete random variable with k possible values, x1 , x2 , . . . , xk . Let pi = P (X = xi ). The variance of X is given by
k 2 = X i=1 k

pi (xi X )2 =
i=1

pi x2 i

2 X

or Var(X ) = E [(X E (X ))2 ] = E (X 2 )[E (X )]2 The standard deviation of X is given by X = SD(X ) = Var(X )

If X is a continuous random variable with density curve f (x),

(xX ) f (x)dx =

x2 f (x)dx2 X

18

Rules for Variances

Var(aX ) = a2 Var(X) Var(X + b) = Var(X) Var(aX + b) = ? Var(X +Y ) = Var(X )+Var(Y )+2 Cov(X, Y ) Var(aX + bY + c) = ? where Cov(X, Y) is the covariance of X and Y .

19

Joint, Marginal, and Conditional Distributions

The conditional distribution of one variable, say X , given the other, say Y , takes a specic value is P (X = xi |Y = yj )

20

Conditional Probability
In general, if P (B ) = 0, the conditional probability of event A given B has occurred, denoted by P (A|B ), is given by P (A B ) P (A|B ) = P (B )

21

Covariance The quantity Cov(X, Y ), called the Covariance of random variables X and Y , is equal to Cov(X, Y ) = E [(X E (X ))(Y E (Y ))] = E (XY ) E (X )E (Y )

If X and Y are discrete random variables, Cov(X, Y ) =

pij (xi X )(yj Y ) pij xi yj

X Y

where pij = P (X = xi , Y = yj ).

22

Correlation The correlation between X and Y is Cov(X, Y ) = Var(X ) Var(Y ) is between -1 and 1 and measures the strength of the linear relationship between X and Y . If X and Y are independent random variables, Cov(X, Y ) = 0, In that case, Var(X + Y ) = Var(X ) + Var(Y ) Var(aX + bY + c) = a2 Var(X ) + b2 Var(Y ) However, Cov(X, Y ) = 0 X and Y are independent.
23

=0

Example
Consider the following joint probability table
X = the average number of dates a person has in a month Y = their annual income in thousands of dollars

Y X 0 1 2 10 .12 .1 .05 20 .12 .25 .2 30 .08 .05 .03

Fill in the marginal distribution of X and Y in the table. Calculate the mean and the variance of both X and Y . Whats the covariance and correlation of X and Y . Are X and Y independent? Why? Find the conditional distribution of X given that Y = 20.
24

The Binomial Distribution The binomial distribution is a very common discrete probability distribution that arises in the following situation: A xed number, n, of trials The n trials are independent of each other Each trial has exactly two outcomes: success and failure The probability of a success, p, is the same for each trial If X is the total number of successes in a binomial setting, then we say that the probability distribution of X is a binomial distribution with parameters n and p: X B (n, p)
25

Examples number of 1s in 10 rolls of a die number of correct guesses on a multiple-choice exam (with 25 questions and 4 choices each) number of Democratic voters in a sample of 3000 people; Binomial or not? X = number of stocks on the NY stock exchange which increase in price on Oct 20, 2003 X = number of games the Bears will win next season X = number of girls among the next 50 children born in Cook County hospital
26

 A couple decides to have children until they have a girl. X = number of boys the couple will have

27

Binomial probability
If X B (n, p), n k P (X = k ) = p (1 p)nk k k = 0 , 1, 2, . . . , n
Why? Since the trials are independent, any particular sequence of 0s and 1s with k 1s and (n k) 0s has probability pk (1 p)nk of happening. There are (n k) 0s.


n k

distinct sequences with k 1s and

Mean and Variance of a Binomial

28

Example Sarah drives to work everyday, but she doesnt want to buy a parking permit. She decides to take her chances and risk getting a parking ticket each day. Suppose 1. A parking permit for a week (5 days) costs $ 50. 2. A parking ne costs $ 30. 3. The probability of getting a parking ticket each day is 0.3. 4. Her chances of getting a ticket each day is independent of other days. 5. She can get only 1 ticket per day. What is her probability of getting at least 1 parking ticket in 1 week (5 days)? What is the expected number of parking tickets that Sarah will get per week?
29