Scribd
Upload
39 views

Conditional Expected Values Independent Random Variables: Stm2Pm

This document summarizes key concepts about conditional expected values and independence of random variables: 1) Conditional expected values are calculated using conditional probability densities or masses. 2) Two examples are given to illustrate calculating conditional expected values for discrete and continuous cases. 3) Two random variables are independent if their joint probability equals the product of their marginal probabilities. 4) Properties of independent random variables are discussed, including that functions of independent variables are also independent, and formulas involving expectation and variance hold for independent variables. 5) The mean and variance of the binomial distribution are summarized.

Uploaded by

ClintonM.Orenge
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
39 views

Conditional Expected Values Independent Random Variables: Stm2Pm

This document summarizes key concepts about conditional expected values and independence of random variables: 1) Conditional expected values are calculated using conditional probability densities or masses. 2) Two examples are given to illustrate calculating conditional expected values for discrete and continuous cases. 3) Two random variables are independent if their joint probability equals the product of their marginal probabilities. 4) Properties of independent random variables are discussed, including that functions of independent variables are also independent, and formulas involving expectation and variance hold for independent variables. 5) The mean and variance of the binomial distribution are summarized.

Uploaded by

ClintonM.Orenge
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

STM2PM

Lecture 28

Conditional Expected Values


Independent Random Variables

28.1
Conditional Expected Values

Definition: Conditional expected values are


calculated using the conditional probability den-
sity (or mass) function:
Z ∞
E(X|Y = y) = x fX|Y (x|y)dx
−∞
and for discrete distributions:
X
E(X|Y = yj ) = xi pX|Y (xi|yj )
xi

28.2
Conditional Expected Values, Example,
Discrete case

Example (see Lecture 27) A fair coin is tossed


three times. Let X be the number of heads on
the first toss, and let Y be the total number of
heads (in the three tosses). From the lecture
27 we know the joint probability mass function
and the marginal probabilities are as follows:
y
0 1 2 3
x 0 1/8 2/8 1/8 0 1/2
1 0 1/8 2/8 1/8 1/2
1/8 3/8 3/8 1/8

We also know that pX|Y (1|2) = 2/3 and


pX|Y (0|2) = 1/3.

Let us calculate E(X|Y = 2):

28.3
Conditional Expected Values, Example,
Continuous case

Example: For the continuous example form


the lecture 27 we have:
X and Y have a joint probability density func-
tion given by

4xy x ∈ [0, 1], y ∈ [0, 1]
f (x, y) =
0 elsewhere
And

2x x ∈ [0, 1], y ∈ [0, 1]
fX|Y (x|y) =
0 elsewhere

From this:
Z 1
E(X|Y = y) = xfX|Y (x|y)dx =
0

28.4
Independence of two random variables,
Introduction

Recall: For two events (called A and B), we


defined independence (in Lecture 6) as fol-
lows:
P (A ∩ B) = P (A)·P (B)
so that
P (A ∩ B)
P (A|B) = = P (A)
P (B)
i.e. being given that B has happened does not
change the probability one assigns to A.

Recall: We know that for a random variable


an ‘event’ means that the variable X takes a
particular value (discrete)

P (X = a)
or its value comes from a set of values (dis-
crete or continuous)

P (a 6 X 6 b)

28.5
Independence of two discrete random
variables

Definition: Two discrete random variables


defined on the same sample space are inde-
pendent if

P (X = x, Y = y) = P (X = x)P (Y = y) (∗)
and in terms of the joint and marginal prob-
ability mass functions

pXY (x, y) = pX (x)pY (y)

To show that two discrete random variables


are not independent, we only need to find one
value of X and one value of Y for which this
condition fails (one counterexample).

Example: In the previous discrete example


(Slide 2)
2
P (X = 1, Y = 2) = pXY (1, 2) = .
8
However, pX (1) = 1
2 and p Y (2) = 3,
8
so we see that
28.6
Independent physical causes

Remark: When it is clear that random vari-


ables arise from independent physical causes,
they will be independent.

In the previous example, we could not expect


independence without checking, as the descrip-
tions of the events do not describe independent
processes.

Exercise: A fair coin is tossed twice. Let


X be the number of heads on the first toss and
Z be the number of tails on the second toss.
Are X and Z independent random variables?

Generally, however, independence depends on


the details of the probability function assigned,
so the condition (∗) must be checked.
28.7
Independence of two continuous random
variables

Definition: Jointly continuous random vari-


ables X and Y are independent if and only if

P (a 6 X 6 b, c 6 Y 6 d) =
P (a 6 X 6 b)P (c 6 Y 6 d)
In terms of the joint and marginal cdfs:

F (x, y) = P (X ≤ x, Y ≤ y)
= P (X ≤ x)P (Y ≤ y)
= FX (x)FY (y)
Or In terms of the joint and marginal pdfs:

f (x, y) = fX (x)fY (y)


.

28.8
Independence of two continuous random
variables, Example

Example: For the previous continuous exam-


ple (Slide 4) in the Lecture 27 we found that
for

4xy x ∈ [0, 1], y ∈ [0, 1]
f (x, y) =
0 otherwise
the marginal probability density functions are

fX (x) = 2x x ∈ [0, 1]
and
fY (y) = 2y y ∈ [0, 1].

That means the random variables X and Y are


independent.

28.9
More examples

Exercise: The function



 2 (x + 2y) 0 6 x 6 1, 0 6 y 6 1
f (x, y) = 3
0 otherwise
is a joint probability distribution function for X
and Y .
Are the variables X and Y independent?

28.10
Properties of independent Random
Variables

Property 1: If X and Y are independent,


then for functions g and h, g(X) and h(Y ) are
independent.

Property 2: Independence can be extended to


more than two random variables X1, X2, . . . Xm
in an obvious way. Then functions of disjoint
subsets of the full set of independent variables
are also independent.

28.11
Properties of independent Random
Variables

Property 3: If X and Y are independent,


then it is valid:

E(XY ) = E(X)E(Y ).
Proof in continuous case:

Remark: The property

E(X + Y ) = E(X) + E(Y )


is valid for any two random variables X and
Y (independence is not necessary)

28.12
Properties of independent Random
Variables

Property 4: If X and Y are independent,


then it is valid:

E((X − µX )(Y − µY )) = 0
Proof:

28.13
Properties of independent Random
Variables

Property 5: If X and Y are independent,


then it is valid:

Var(X + Y ) = Var(X) + Var(Y )


Proof:

28.14
Mean and Variance for the Binomial
Distribution.

Let X1, X2, . . . , Xn be independent Bernoulli ran-


dom variables, each with the same success prob-
ability p.

What is the distribution of


Y = X1 + X2 + . . . + Xn?

From Lecture 23 we know that for each Xi,


E(Xi) = p.
Hence the mean of the binomial distribution
Bin(n, p) is:

We also found Var(Xi) = p(1 − p)

Hence the variance of the binomial distribution


is:

28.15

You might also like