Scribd Logo
Upload

Welcome to Scribd!

  • 36 views

    Manipulating Continuous Random Variables Class 5, 18.05, Spring 2014 Jeremy Orloff and Jonathan Bloom 1 Learning Goals

    Uploaded by

    Pavan Velivela
    This document discusses manipulating continuous random variables through transformations. It provides four examples: 1) If X is uniformly distributed from 0 to 2, Y=X^2 is uniformly distributed from 0 to 4 with a probability density function of 1/√(4y). 2) If X is exponentially distributed with rate λ, Y=X^2 has a probability density function of √(λ/2y)e^(-λ√y). 3) If X is normally distributed with mean 5 and variance 9, Z=(X-5)/3 is standard normal. 4) More generally, if X is normally distributed with mean μ and variance σ^2, Z=(X-

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Manipulating Continuous Random Variables Class 5, 18.05, Spring 2014 Jeremy Orloff and Jonathan Bloom 1 Learning Goals

    Uploaded by

    Pavan Velivela
    36 views3 pages
    This document discusses manipulating continuous random variables through transformations. It provides four examples: 1) If X is uniformly distributed from 0 to 2, Y=X^2 is uniformly distributed from 0 to 4 with a probability density function of 1/√(4y). 2) If X is exponentially distributed with rate λ, Y=X^2 has a probability density function of √(λ/2y)e^(-λ√y). 3) If X is normally distributed with mean 5 and variance 9, Z=(X-5)/3 is standard normal. 4) More generally, if X is normally distributed with mean μ and variance σ^2, Z=(X-

    Original Title

    Lecture probability

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses manipulating continuous random variables through transformations. It provides four examples: 1) If X is uniformly distributed from 0 to 2, Y=X^2 is uniformly distributed from 0 to 4 with a probability density function of 1/√(4y). 2) If X is exponentially distributed with rate λ, Y=X^2 has a probability density function of √(λ/2y)e^(-λ√y). 3) If X is normally distributed with mean 5 and variance 9, Z=(X-5)/3 is standard normal. 4) More generally, if X is normally distributed with mean μ and variance σ^2, Z=(X-

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    36 views3 pages

    Manipulating Continuous Random Variables Class 5, 18.05, Spring 2014 Jeremy Orloff and Jonathan Bloom 1 Learning Goals

    Uploaded by

    Pavan Velivela
    This document discusses manipulating continuous random variables through transformations. It provides four examples: 1) If X is uniformly distributed from 0 to 2, Y=X^2 is uniformly distributed from 0 to 4 with a probability density function of 1/√(4y). 2) If X is exponentially distributed with rate λ, Y=X^2 has a probability density function of √(λ/2y)e^(-λ√y). 3) If X is normally distributed with mean 5 and variance 9, Z=(X-5)/3 is standard normal. 4) More generally, if X is normally distributed with mean μ and variance σ^2, Z=(X-

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    Manipulating Continuous Random Variables

    Class 5, 18.05, Spring 2014

    Jeremy Orloff and Jonathan Bloom

    1 Learning Goals
    1. Be able to find the pdf and cdf of a random variable defined in terms of a random variable
    with known pdf and cdf.

    2 Transformations of Random Variables

    If Y = aX +b then the properties of expectation and variance tell us that E(Y ) = aE(X)+b
    and Var(Y ) = a2 Var(X). But what is the distribution function of Y ? If Y is continuous,
    what is its pdf?
    Often, when looking at transforms of discrete random variables we work with tables.
    For continuous random variables transforming the pdf is just change of variables (‘u­
    substitution’) from calculus. Transforming the cdf makes direct use of the definition of
    the cdf.
    Let’s remind ourselves of the basics:
    1. The cdf of X is FX (x) = P (X ≤ x).
    2. The pdf of X is related to FX by fX (x) = FX' (x).

    Example 1. Let X ∼ U (0, 2), so fX (x) = 1/2 and FX (x) = x/2 on [0,2]. What is the
    range, pdf and cdf of Y = X 2 ?

    answer: The range is easy: [0, 4].

    To find the cdf we work systematically from the definition.

    √ √ √
    FY (y) = P (Y ≤ y) = P (X 2 ≤ y) = P (X ≤ y) = FX ( y) = y/2.

    To find the pdf we can just differentiate the cdf

    d 1
    fY (y) = FY (y) = √ .
    dy 4 y

    An alternative way to find the pdf directly is by change of variables. The trick here is to
    remember that it is fX (x)dx which gives probability (fX (x) by itself is probability density).
    Here is how the calculation goes in this example.
    dy
    y = x2 ⇒ dy = 2x dx ⇒ dx = √
    2 y
    dx dy
    fX (x) dx = = √ = fY (y) dy
    2 4 y

    dy
    Therefore fY (y) = √
    4 y

    18.05 class 5, Manipulating Continuous Random Variables, Spring 2014 2

    Example 2. Let X ∼ exp(λ), so fX (x) = λe−λx on [0, ∞]. What is the density of Y = X 2 ?
    answer: Let’s do this using the change of variables.
    dy
    y = x2 ⇒ dy = 2x dx ⇒ dx = √
    2 y
    √ dy
    fX (x) dx = λe−λx dx = λe−λ y √ = fY (y) dy
    2 y

    λ √
    Therefore fY (y) = √ e−λ y .
    2 y

    X −5
    Example 3. Assume X ∼ N(5, 32 ). Show that Z = is standard normal, i.e.,
    3
    Z ∼ N(0, 1).
    answer: Again using the change of variables and the formula for fX (x) we have

    x−5 dx
    z= ⇒ dz = ⇒ dx = 3 dz
    3 3
    1 2 2 1 2 1 2
    fX (x) dx = √ e−(x−5) /(2·3 ) dx = √ e−z /2 3 dz = √ e−z /2 dz = fZ (z) dz
    3 2π 3 2π 2π
    1 2
    Therefore fZ (z) = √ e−z /2 . Since this is exactly the density for N(0, 1) we have shown
    2π
    that Z is standard normal.

    This example shows an important general property of normal random variables which we
    give in the next example.
    X −μ
    Example 4. Assume X ∼ N(μ, σ 2 ). Show that Z = is standard normal, i.e.,
    σ
    Z ∼ N(0, 1).
    answer: This is exactly the same computation as the previous example with μ replacing 5
    and σ replacing 3. We show the computation without comment.
    x−μ dx
    z= ⇒ dz = ⇒ dx = σ dz
    σ σ
    1 2 2 1 2 1 2
    fX (x) dx = √ e−(x−μ) /(2·σ ) dx = √ e−z /2 σ dz = √ e−z /2 dz = fZ (z) dz
    σ 2π σ 2π 2π
    1 2
    Therefore fZ (z) = √ e−z /2 . This shows Z is standard normal.
    2π
    MIT OpenCourseWare
    http://ocw.mit.edu

    18.05 Introduction to Probability and Statistics


    Spring 2014

    For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

    You might also like