Upload
shivam-taji
View
213
Download
0
Embed Size (px)
Citation preview
7/25/2019 Module_13_0
1/16
Module 13
Transformations of Absolutely Continuous
Random Variables
() Module 13 Transformations of Absolutely Continuous Random Variables
1 / 15
http://find/7/25/2019 Module_13_0
2/16
X: an A.C. r.v. with d.f. FX() and p.d.f. fX();
FX(x) = x
fX(t)dt, x R;
For
7/25/2019 Module_13_0
3/16
We will assume throughout that fX() is continuous everywhere except
at (possibly) finite number of points (say, x1, x2, . . . , xn), where it hasjump discontinuities. In that case FX() is differentiable everywhereexcept at discontinuity points{x1, . . . , xn} offX(). Moreover
fX(t) =
FX
(t), ift / {
x1
, . . . , xn}
0, otherwise.
Support
SX = {x R :FX(x+) FX(x )>0, >0} .
() Module 13 Transformations of Absolutely Continuous Random Variables
2 / 15
http://find/http://goback/7/25/2019 Module_13_0
4/16
g :R
R
: a given function;
Then Y =g(X) is a r.v.;
Goal: To find the probability distribution (i.e., d.f. FY() and/or,p.d.f./p.m.f. fY()) ofY =g(X);
Remark 1: We have seen that when X is discrete, Y =g(X) is also
discrete. When X is A.C., Y =g(X) may not be A.C. (or evencontinuous) as the following example illustrates.
() Module 13 Transformations of Absolutely Continuous Random Variables 3 / 15
http://find/7/25/2019 Module_13_0
5/16
Example 1: Let Xbe an A.C. r.v. with p.d.f.
fX(x) =
12 , if 1< x
7/25/2019 Module_13_0
6/16
P({
Y = 0}
) = P({
0
X
7/25/2019 Module_13_0
7/16
Result 1: Suppose SX =k
i=1Si,X, where{Si,X, i= 1, . . . , k} is acollection of disjoint intervals and in Si,X (i= 1, ..., k), g :Si,X
R is
strictly monotone with inverse function g1i (y) such that ddy
g1i (y) iscontinuous. Let g(Si,X) = {g(x) :x Si,X}, i= 1, ..., k. Then the r.v.Y =g(X) is A.C. with p.d.f.
fY(y) =
ki=1
fX
g1i (y) ddyg1i (y)
Ig(Si,X)(y),
where, for a set A, IA() denotes its indicator function, i.e.,
IA(y) =
1, ify A0, otherwise
.
() Module 13 Transformations of Absolutely Continuous Random Variables 6 / 15
http://find/7/25/2019 Module_13_0
8/16
Corollary 1: Suppose that g :SX R is strictly monotone with inversefunction g1(y) such that d
dyg1(y) is continuous. Let
g(SX) = {g(x) :x SX}. Then Y =g(X) is of A.C. type with p.d.f.
fY(y) =fX(g1(y))
ddyg1(y) Ig(SX)(y).
() Module 13 Transformations of Absolutely Continuous Random Variables 7 / 15
http://find/7/25/2019 Module_13_0
9/16
Example 2: Let X be an A.C. r.v. with p.d.f.
fX(x) =
|x|2 , if 1
7/25/2019 Module_13_0
10/16
S1,X = [1, 0) S2,X = [0, 2]g11 (y) =
y g12 (y) =
y
ddy
g11 (y) = 12y
ddy
g12 (y) = 12y
g(S1,X) = (0, 1] g(S2,X) = [0, 4)y g(S1,X) 0< y 1 y g(S2,X) 0 y 4fX(g
11 (y))
ddyg11 (y) fX(g12 (y))
ddyg12 (y)
=fX(
y)
12y I(0,1](y) =fX(
y)
12y I[0,4](y)
Thus a p.d.f. ofY is
fY(y) =2
i=1
fX(g1i (y))
d
dyg1i (y)
Ig(Si,X)(y)
=
12 , if 0
7/25/2019 Module_13_0
11/16
FY(y) =P({Y y}) = y
fY(t)dt=
0, ify
7/25/2019 Module_13_0
12/16
(b) For y
7/25/2019 Module_13_0
13/16
For y 4, FY(y) = 1.
Thus the d.f. of Y is
FY(y) =
0, ify
7/25/2019 Module_13_0
14/16
ClearlySY = [0, 4], FY() is differentiable everywhere except at points 0, 1and 4. Let
g(y) = FY(y), ify / {0, 1, 4}
0, otherwise
=
12 , if 0
7/25/2019 Module_13_0
15/16
Take home problem
Let X be a r.v. with p.d.f.
fX(x) = ex, ifx>0
0, otherwise .
Let Y = 2X+ 3.
(a) Find the p.d.f. of Y and hence find the d.f. of Y;
(b) Find the d.f. of Y and hence find the p.d.f. of Y.
() Module 13 Transformations of Absolutely Continuous Random Variables 14 / 15
http://find/7/25/2019 Module_13_0
16/16
Abstract of Next Module
X: a r.v. associated with a random experiment
E;
Each time the random experiment is performed we get a value ofX;
Question: If the random experiment is performed infinitely what is themean (or expectation) of observed values ofX org(X), for some functionreal-valued function g()?In the discrete case, the relative frequency interpretation of probabilitysuggests that we take
E(g(X)) = limN
xSX g(x) Number of times we get {X =x}
N
=xSX
g(x) limN
frequency of{X =x}N
=xSX
g(x)P({X =x}) =xSX
g(x)fX(x).
() Module 13 Transformations of Absolutely Continuous Random Variables 15 / 15
http://find/