Some Basic Probability Concepts. 2.1 Experiments, Outcomes and Random Variables

Similar documents
Econ 325: Introduction to Empirical Economics

Probability 2 - Notes 10. Lemma. If X is a random variable and g(x) 0 for all x in the support of f X, then P(g(X) 1) E[g(X)].

Random Variables, Sampling and Estimation

Chapter 2. Some Basic Probability Concepts. 2.1 Experiments, Outcomes and Random Variables

Quick Review of Probability

Quick Review of Probability

CEE 522 Autumn Uncertainty Concepts for Geotechnical Engineering

Joint Probability Distributions and Random Samples. Jointly Distributed Random Variables. Chapter { }

It is always the case that unions, intersections, complements, and set differences are preserved by the inverse image of a function.

STATISTICAL METHODS FOR BUSINESS

Distribution of Random Samples & Limit theorems

EE 4TM4: Digital Communications II Probability Theory

17. Joint distributions of extreme order statistics Lehmann 5.1; Ferguson 15

Lecture 7: Properties of Random Samples

IIT JAM Mathematical Statistics (MS) 2006 SECTION A

( ) = p and P( i = b) = q.

TAMS24: Notations and Formulas

Probability and statistics: basic terms

Exponential Families and Bayesian Inference

APPLIED MULTIVARIATE ANALYSIS

NOTES ON DISTRIBUTIONS

AMS570 Lecture Notes #2

Lecture 1 Probability and Statistics

2. The volume of the solid of revolution generated by revolving the area bounded by the

Probability and Statistics

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 6 9/24/2008 DISCRETE RANDOM VARIABLES AND THEIR EXPECTATIONS

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam.

Convergence of random variables. (telegram style notes) P.J.C. Spreij

The standard deviation of the mean

Expectation and Variance of a random variable

Unbiased Estimation. February 7-12, 2008

Final Examination Statistics 200C. T. Ferguson June 10, 2010

Lecture 11 and 12: Basic estimation theory

First Year Quantitative Comp Exam Spring, Part I - 203A. f X (x) = 0 otherwise

Limit Theorems. Convergence in Probability. Let X be the number of heads observed in n tosses. Then, E[X] = np and Var[X] = np(1-p).

CH.25 Discrete Random Variables

Lecture 1 Probability and Statistics

We are mainly going to be concerned with power series in x, such as. (x)} converges - that is, lims N n

Lecture 3. Properties of Summary Statistics: Sampling Distribution

4. Partial Sums and the Central Limit Theorem

CS166 Handout 02 Spring 2018 April 3, 2018 Mathematical Terms and Identities

Probability and Random Processes

Generalized Semi- Markov Processes (GSMP)

CONTENTS. Course Goals. Course Materials Lecture Notes:

The variance of a sum of independent variables is the sum of their variances, since covariances are zero. Therefore. V (xi )= n n 2 σ2 = σ2.

Direction: This test is worth 150 points. You are required to complete this test within 55 minutes.

Topic 9: Sampling Distributions of Estimators

Lecture 12: September 27

Maximum Likelihood Estimation

[ ] ( ) ( ) [ ] ( ) 1 [ ] [ ] Sums of Random Variables Y = a 1 X 1 + a 2 X 2 + +a n X n The expected value of Y is:

CH5. Discrete Probability Distributions

The Poisson Process *

Formula Sheet. December 8, 2011

Matrix Representation of Data in Experiment

Binomial Distribution

Economics 241B Relation to Method of Moments and Maximum Likelihood OLSE as a Maximum Likelihood Estimator

Unit 6: Sequences and Series

Lecture 2: Poisson Sta*s*cs Probability Density Func*ons Expecta*on and Variance Es*mators

Lecture 8: Convergence of transformations and law of large numbers

Lecture 33: Bootstrap

MASSACHUSETTS INSTITUTE OF TECHNOLOGY 6.436J/15.085J Fall 2008 Lecture 19 11/17/2008 LAWS OF LARGE NUMBERS II THE STRONG LAW OF LARGE NUMBERS

EECS564 Estimation, Filtering, and Detection Hwk 2 Solns. Winter p θ (z) = (2θz + 1 θ), 0 z 1

6. Sufficient, Complete, and Ancillary Statistics

NANYANG TECHNOLOGICAL UNIVERSITY SYLLABUS FOR ENTRANCE EXAMINATION FOR INTERNATIONAL STUDENTS AO-LEVEL MATHEMATICS

Mathematical Statistics - MS

ECONOMETRIC THEORY. MODULE XIII Lecture - 34 Asymptotic Theory and Stochastic Regressors

Lecture 18: Sampling distributions

5. INEQUALITIES, LIMIT THEOREMS AND GEOMETRIC PROBABILITY

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Discrete Mathematics and Probability Theory Spring 2016 Rao and Walrand Note 19

Lecture 20: Multivariate convergence and the Central Limit Theorem

Large Sample Theory. Convergence. Central Limit Theorems Asymptotic Distribution Delta Method. Convergence in Probability Convergence in Distribution

Lecture 4. Random variable and distribution of probability

Topic 8: Expected Values

Probability and Distributions. A Brief Introduction

Chapter 5. Inequalities. 5.1 The Markov and Chebyshev inequalities

Sample Correlation. Mathematics 47: Lecture 5. Dan Sloughter. Furman University. March 10, 2006

Sequences, Mathematical Induction, and Recursion. CSE 2353 Discrete Computational Structures Spring 2018

This section is optional.

Chapter 3. Strong convergence. 3.1 Definition of almost sure convergence

Summary. Recap ... Last Lecture. Summary. Theorem

Learning Theory: Lecture Notes

The Central Limit Theorem

Introduction to Probability I: Expectations, Bayes Theorem, Gaussians, and the Poisson Distribution. 1

f X (12) = Pr(X = 12) = Pr({(6, 6)}) = 1/36

Discrete probability distributions

Machine Learning Brett Bernstein

1.010 Uncertainty in Engineering Fall 2008

Lecture 5. Random variable and distribution of probability

ST5215: Advanced Statistical Theory

Describing the Relation between Two Variables

[ 11 ] z of degree 2 as both degree 2 each. The degree of a polynomial in n variables is the maximum of the degrees of its terms.

Approximations and more PMFs and PDFs


HOMEWORK I: PREREQUISITES FROM MATH 727

ACE 562 Fall Lecture 2: Probability, Random Variables and Distributions. by Professor Scott H. Irwin

Chapter 6 Sampling Distributions

STAT Homework 1 - Solutions

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Transcription:

Some Basic Probability Cocepts 2. Experimets, Outcomes ad Radom Variables A radom variable is a variable whose value is ukow util it is observed. The value of a radom variable results from a experimet; it is ot perfectly predictable. A discrete radom variable ca take oly a fiite umber of values, that ca be couted by usig the positive itegers. A cotiuous radom variable ca take ay real value (ot just whole umbers) i a iterval o the real umber lie. Slide 2.

2.2 The Probability Distributio of a Radom Variable Whe the values of a discrete radom variable are listed with their chaces of occurrig, the resultig table of outcomes is called a probability fuctio or a probability desity fuctio. For a discrete radom variable X the value of the probability desity fuctio f(x) is the probability that the radom variable X takes the value x, f(x)=p(x=x). Therefore, 0 f(x) ad, if X takes values x,.., x, the f( x) + f( x2) + L+ f( x ) =. For the cotiuous radom variable Y the probability desity fuctio f(y) ca be represeted by a equatio, which ca be described graphically by a curve. For cotiuous radom variables the area uder the probability desity fuctio correspods to probability. Slide 2.2

2.3 Expected Values Ivolvig a Sigle Radom Variable 2.3. The Rules of Summatio. If X takes values x,..., x the their sum is 2. If a is a costat, the i 2 L i= x = x + x + + x 3. If a is a costat the i= a = a ax = a x i i= i= i Slide 2.3

4. If X ad Y are two variables, the 5. If X ad Y are two variables, the ( x + y ) = x + y i i i i i= i= i= ( ax + by ) = a x + b y i i i i i= i= i= 6. The arithmetic mea (average) of values of X is Also, x xi i= x+ x2 + L + x. = = i= ( x x) = 0 i Slide 2.4

7. We ofte use a abbreviated form of the summatio otatio. For example, if f(x) is a fuctio of the values of X, i= f( x ) = f( x ) + f( x ) + L+ f( x ) i 2 = f ( x ) ("Sum over all values of the idex i") = i x i f ( x) ("Sum over all possible values of X") 8. Several summatio sigs ca be used i oe expressio. Suppose the variable Y takes values ad X takes m values, ad let f(x,y)=x+y. The the double summatio of this fuctio is m m f ( x, y ) = ( x + y j ) i j i i= j= i= j= Slide 2.5

To evaluate such expressios work from the iermost sum outward. First set i= ad sum over all values of j, ad so o. That is, To illustrate, let m = 2 ad = 3. The ( i, j) = ( i, ) + ( i, 2) + ( i, 3) 2 3 2 f x y f x y f x y f x y i= j= i= (, ) (, 2) (, 3) (, ) + (, ) + (, ) = f x y + f x y + f x y + f x y f x y f x y 2 2 2 2 3 The order of summatio does ot matter, so m m f ( x, y ) = f( x, y ) i j i j i= j= j= i= Slide 2.6

2.3.2 The Mea of a Radom Variable The expected value of a radom variable X is the average value of the radom variable i a ifiite umber of repetitios of the experimet (repeated samples); it is deoted E[X]. If X is a discrete radom variable which ca take the values x, x2,,x with probability desity values f(x ), f(x2),, f(x), the expected value of X is EX [ ] = xf( x) + xf( x) + L+ xf( x) = = 2 2 i= x xf( x) i xf ( x) i (2.3.) Slide 2.7

2.3.3 Expectatio of a Fuctio of a Radom Variable If X is a discrete radom variable ad g(x) is a fuctio of it, the EgX [ ( )] = gx ( ) f( x) (2.3.2a) x However, EgX [ ( )] gex [ ( )] i geeral. Slide 2.8

If X is a discrete radom variable ad g(x) = g (X) + g 2 (X), where g (X) ad g 2 (X) are fuctios of X, the EgX [ ( )] = [ g( x) + g( x)] f( x) x 2 (2.3.2b) = g ( x) f( x) + g ( x) f( x) x 2 x = Eg [ ( x)] + Eg [ ( x)] 2 The expected value of a sum of fuctios of radom variables, or the expected value of a sum of radom variables, is always the sum of the expected values. If c is a costat, Ec [] = c (2.3.3a) If c is a costat ad X is a radom variable, the Slide 2.9

EcX [ ] = cex [ ] (2.3.3b) If a ad c are costats the Ea [ + cx] = a+ cex [ ] (2.3.3c) 2.3.4 The Variace of a Radom Variable [ ] 2 var( X ) EgX [ ( )] E X EX ( ) EX [ ] [ EX ( )] 2 2 2 =σ = = = (2.3.4) Let a ad c be costats, ad let Z = a + cx. The Z is a radom variable ad its variace is 2 2 var( a+ cx) = E[( a+ cx) E( a+ cx)] = c var( X) (2.3.5) Slide 2.0

2.4 Usig Joit Probability Desity Fuctios 2.4. Margial Probability Desity Fuctios If X ad Y are two discrete radom variables the f( x) = f( x, y) for each value X ca take y f( y) = f( x, y) for each value Y ca take x (2.4.) 2.4.2 Coditioal Probability Desity Fuctios f ( xy, ) f( x y) = P[ X = x Y = y] = (2.4.2) f ( y) Slide 2.

2.4.3 Idepedet Radom Variables If X ad Y are idepedet radom variables, the f ( xy, ) = f( xf ) ( y) (2.4.3) for each ad every pair of values x ad y. The coverse is also true. If X,, X are statistically idepedet the joit probability desity fuctio ca be factored ad writte as f 2 2 2 ( x, x, L, x ) = f ( x ) f ( x ) K f ( x ) (2.4.4) Slide 2.2

If X ad Y are idepedet radom variables, the the coditioal probability desity fuctio of X give that Y=y is f( x, y) f( x) f( y) f ( x y) = f( x) f( y) = f( y) = (2.4.5) for each ad every pair of values x ad y. The coverse is also true. 2.5 The Expected Value of a Fuctio of Several Radom Variables: Covariace ad Correlatio If X ad Y are radom variables, the their covariace is cov( X, Y) = E[( X E[ X])( Y E[ Y])] (2.5.) Slide 2.3

If X ad Y are discrete radom variables, f(x,y) is their joit probability desity fuctio, ad g(x,y) is a fuctio of them, the EgXY [ (, )] = gxy (, ) f( xy, ) (2.5.2) x y If X ad Y are discrete radom variables ad f(x,y) is their joit probability desity fuctio, the cov( X, Y) = E[( X E[ X])( Y E[ Y])] = [ x E( X)][ y E( Y)] f( x, y) x y (2.5.3) Slide 2.4

If X ad Y are radom variables the their correlatio is cov( XY, ) ρ = (2.5.4) var( X ) var( Y ) 2.5. The Mea of a Weighted Sum of Radom Variables EaX [ + by] = ae( X) + bey ( ) (2.5.5) If X ad Y are radom variables, the E[ X + Y] = E[ X] + E[ Y] (2.5.6) Slide 2.5

2.5.2 The Variace of a Weighted Sum of Radom Variables If X, Y, ad Z are radom variables ad a, b, ad c are costats, the 2 2 2 [ ax + by + cz ] = a [ X ] + b [ Y ] + c [ Z ] + 2abcov [ X, Y ] + 2accov [ X, Z ] + 2bccov [ Y, Z ] var var var var (2.5.7) If X, Y, ad Z are idepedet, or ucorrelated, radom variables, the the covariace terms are zero ad: [ ax by cz ] a 2 [ X ] b 2 [ Y ] c 2 [ Z ] var + + = var + var + var (2.5.8) Slide 2.6

If X, Y, ad Z are idepedet, or ucorrelated, radom variables, ad if a = b = c =, the [ X Y Z] [ X] [ Y] [ Z] var + + = var + var + var (2.5.9) 2.6 The Normal Distributio 2 ( x β) f( x) = exp, < x< 2 2σ 2 2 πσ (2.6.) Z = X β σ ~ N(0,) Slide 2.7

If X ~ N(β, 2 σ ) ad a is a costat, the X a a PX [ a] P β β P Z β = = σ σ σ (2.6.2) If X ~ N(β, 2 σ ) ad a ad b are costats, the a X b a b Pa [ X b] P β β β P β Z β = = σ σ σ σ σ (2.6.3) If X ~ N(, 2 ), X ~ N(, 2 ), X ~ N(, 2 ) β σ 2 β2 σ2 3 β3 σ 3 ad c, c, c3 are costats, the ( ) ( ) Z = cx + ~,var cx + 2 2 cx 3 3 N E Z Z (2.6.4) 2 Slide 2.8