Econ 371 Problem Set #1 Answer Sheet

Similar documents
Business Statistics 41000: Homework # 2 Solutions

Raquel Prado. Name: Department of Applied Mathematics and Statistics AMS-131. Spring 2010

Properties of Summation Operator

ECE 302, Final 3:20-5:20pm Mon. May 1, WTHR 160 or WTHR 172.

Lecture 13 (Part 2): Deviation from mean: Markov s inequality, variance and its properties, Chebyshev s inequality

More than one variable

STA 2201/442 Assignment 2

ECON Fundamentals of Probability

Exam 1 Review With Solutions Instructor: Brian Powers

EXAMINATIONS OF THE HONG KONG STATISTICAL SOCIETY GRADUATE DIPLOMA, Statistical Theory and Methods I. Time Allowed: Three Hours

Lecture 2: Review of Probability

18.05 Exam 1. Table of normal probabilities: The last page of the exam contains a table of standard normal cdf values.

EECS 126 Probability and Random Processes University of California, Berkeley: Spring 2015 Abhay Parekh February 17, 2015.

Probability & Statistics - FALL 2008 FINAL EXAM

1 Probability Distributions

Stat 704 Data Analysis I Probability Review

EXAMINATIONS OF THE ROYAL STATISTICAL SOCIETY

Introduction to Probability Theory for Graduate Economics Fall 2008

Sections 5.1 and 5.2

Quiz 1. Name: Instructions: Closed book, notes, and no electronic devices.

6.041/6.431 Fall 2010 Quiz 2 Solutions

Objective - To understand experimental probability

STOR Lecture 16. Properties of Expectation - I

The probability of an event is viewed as a numerical measure of the chance that the event will occur.

Basic Probability Reference Sheet

Multivariate probability distributions and linear regression

1 Random variables and distributions

UC Berkeley Department of Electrical Engineering and Computer Science. EE 126: Probablity and Random Processes. Problem Set 9 Fall 2007

2. Variance and Covariance: We will now derive some classic properties of variance and covariance. Assume real-valued random variables X and Y.

18.440: Lecture 26 Conditional expectation

Continuous Probability Distributions

Statistics and Sampling distributions

Week 12-13: Discrete Probability

Probability Theory and Statistics. Peter Jochumzen

18.440: Lecture 19 Normal random variables

Math 416 Lecture 3. The average or mean or expected value of x 1, x 2, x 3,..., x n is

Review of probability and statistics 1 / 31

Probability Theory. Introduction to Probability Theory. Principles of Counting Examples. Principles of Counting. Probability spaces.

MATH 151, FINAL EXAM Winter Quarter, 21 March, 2014

E X A M. Probability Theory and Stochastic Processes Date: December 13, 2016 Duration: 4 hours. Number of pages incl.

Chapter 4 : Expectation and Moments

Covariance and Correlation Class 7, Jeremy Orloff and Jonathan Bloom

STT 315 Problem Set #3

IEOR 3106: Introduction to Operations Research: Stochastic Models. Professor Whitt. SOLUTIONS to Homework Assignment 2

Math 180B Problem Set 3

Lecture 2: Repetition of probability theory and statistics

Expectation and Variance

Random variables (discrete)

Introduction. The Linear Regression Model One popular model is the linear regression model. It writes as :

Random Variable. Discrete Random Variable. Continuous Random Variable. Discrete Random Variable. Discrete Probability Distribution

First Midterm Examination Econ 103, Statistics for Economists February 14th, 2017

FINAL EXAM: 3:30-5:30pm

Random Variables and Expectations

Math 151. Rumbos Fall Solutions to Review Problems for Exam 2. Pr(X = 1) = ) = Pr(X = 2) = Pr(X = 3) = p X. (k) =

1. Regressions and Regression Models. 2. Model Example. EEP/IAS Introductory Applied Econometrics Fall Erin Kelley Section Handout 1

Review of Probability. CS1538: Introduction to Simulations

2. Suppose (X, Y ) is a pair of random variables uniformly distributed over the triangle with vertices (0, 0), (2, 0), (2, 1).

4. Suppose that we roll two die and let X be equal to the maximum of the two rolls. Find P (X {1, 3, 5}) and draw the PMF for X.

Statistics for Managers Using Microsoft Excel/SPSS Chapter 4 Basic Probability And Discrete Probability Distributions

FINAL EXAM: Monday 8-10am

M378K In-Class Assignment #1

1 Basic continuous random variable problems

1 Basic continuous random variable problems

Probability and random variables

Recursive Estimation

Statistics for Managers Using Microsoft Excel (3 rd Edition)

MATHEMATICS 154, SPRING 2009 PROBABILITY THEORY Outline #11 (Tail-Sum Theorem, Conditional distribution and expectation)

Some Basic Concepts of Probability and Information Theory: Pt. 2

Mathematics of Finance Problem Set 1 Solutions

IEOR 3106: Introduction to Operations Research: Stochastic Models. Professor Whitt. SOLUTIONS to Homework Assignment 1

Lecture 1: Probability Fundamentals

Probability. Hosung Sohn

Machine Learning: Homework Assignment 2 Solutions

This does not cover everything on the final. Look at the posted practice problems for other topics.

No books, no notes, only SOA-approved calculators. Please put your answers in the spaces provided!

STAT 516 Midterm Exam 3 Friday, April 18, 2008

Discrete Distributions

This exam is closed book and closed notes. (You will have access to a copy of the Table of Common Distributions given in the back of the text.

PHP2510: Principles of Biostatistics & Data Analysis. Lecture X: Hypothesis testing. PHP 2510 Lec 10: Hypothesis testing 1

SCHOOL OF MATHEMATICS AND STATISTICS

Preliminary statistics

[POLS 8500] Review of Linear Algebra, Probability and Information Theory

Mock Exam - 2 hours - use of basic (non-programmable) calculator is allowed - all exercises carry the same marks - exam is strictly individual

Course: ESO-209 Home Work: 1 Instructor: Debasis Kundu

Homework 10 (due December 2, 2009)

Midterm Exam 1 Solution

Probability and random variables. Sept 2018

ECE 450 Homework #3. 1. Given the joint density function f XY (x,y) = 0.5 1<x<2, 2<y< <x<4, 2<y<3 0 else

CSE 103 Homework 8: Solutions November 30, var(x) = np(1 p) = P r( X ) 0.95 P r( X ) 0.

Lecture 10: Bayes' Theorem, Expected Value and Variance Lecturer: Lale Özkahya

Statistics STAT:5100 (22S:193), Fall Sample Final Exam B

Class 26: review for final exam 18.05, Spring 2014

Algorithms for Uncertainty Quantification

. Find E(V ) and var(v ).

Problem Set #6: OLS. Economics 835: Econometrics. Fall 2012

Statistics and Econometrics I

Covariance and Correlation

Part 3: Parametric Models

Review of Basic Probability Theory

Dept. of Linguistics, Indiana University Fall 2015

Transcription:

Econ 371 Problem Set #1 Answer Sheet 2.1 In this question, you are asked to consider the random variable Y, which denotes the number of heads that occur when two coins are tossed. a. The first part of the question asks you to derive the probability distribution for Y. For a discrete random variable, this is just a listing of all the possible outcomes and the probability that they can occur. Assuming that the two coins are fair, there are only four possible results when we flip the two coins: Heads,Heads (Y = 2) Heads,Tails (Y = 1) Tails,Heads (Y = 1) Tails,Tails (Y = 0) which happen with equal probability. The corresponding probability distribution is given by the following table. outcome (number of heads) Y = 0 Y = 1 Y = 2 probability 0.25 0.50 0.25 b. The cumulative probability distribution function for Y corresponds to just computing the probabilities of Y being less than or equal to a given value outcome (number of heads) Y < 0 0 Y < 1 1 Y < 2 Y > 2 cumulative probability 0 0.25 0.75 1 c. Finally, you are asked to derive the mean and variance of Y. These are given by: and E(Y ) = V ar(y ) = E[(Y µ Y ) 2 ] = k P r(y = y i )y i = k p i y i = (0.25) 0 + (0.50) 1 + (0.25) 2 = 1 k p i (y i µ Y ) 2 = (0.25) (0 1) 2 + (0.50) (1 1) 2 + (0.25) (2 1) 2 = (0.25) 1 + (0.50) 0 + (0.25) 1 = 0.5 2.3 In this question, you are asked to compute various characteristics of two random variables, W and V that are functions of the random variables in Table 2.2 (X and Y ). Specifically, we have: W = 3 + 6X V = 20 7Y a. First, you are asked to compute E(W ) and E(V ). From equation (2.12) in the text, we know that: E(W ) = 3 + 6E(X) (1) 1

but, using Table 2.2 and our formula for computing means, we have that E(X) = k X p i x i = (0.30) 0 + (0.70) 1 = 0.7 Substituting this into (1) yields. E(W ) = 3 + 6E(X) = 3 + 6(0.7) = 7.2 (2) Similarly, E(Y ) = k Y p i x i = (0.22) 0 + (0.78) 1 = 0.78, so that E(V ) = 20 7E(Y ) = 20 7(0.78) = 14.54. (3) b. In part b you are asked to compute the corresponding variances for W and V. Equation (2.13) in the text tells us that: σw 2 = 6 2 σx 2 = 36σX 2 σv 2 = ( 7) 2 σy 2 = 49σY 2 Turning to our formulas for variances, however, we know that: V ar(x) = k X p i (x i µ X ) 2 = (0.30) (0 0.7) 2 + (0.70) (1 0.7) 2 = (0.30) (0.49) + (0.70) (0.09) = 0.21 So that Similarly, σ 2 W = 6 2 σ 2 X = 36(0.21) = 7.56 (4) V ar(y ) = k Y p i (y i µ Y ) 2 = (0.22) (0 0.78) 2 + (0.78) (1 0.78) 2 = (0.22) (0.6084) + (0.78) (0.0484) = 0.1716 So that σ 2 V = 49 2 σ 2 Y = 49(0.1716) = 8.4084 (5) c. Finally, you are asked to compute σ W V and corr(w, V ) There are two ways to do this (at least). First, you can derive the formula for the covariance term using: Cov(W, V ) = Cov(3 + 6X, 20 7Y ) = 6( 7)Cov(X, Y ) (6) 2

But, using the formula in equation (2.24) in the text, we have that: So that Cov(X, Y ) = σ XY = E [(X µ X )(Y µ Y )] = k X k Y j=1 x i µ X )(y j µ Y )P r(x = x i, Y = y j ) = (0 0.7)(0 0.78)(0.15) + (1 0.7)(0 0.78)(0.07) + (0 0.7)(1 0.78)(0.15) + (1 0.7)(1 0.78)(0.63) = 0.0819 0.01638 0.0231 + 0.04158 = 0.084 Cov(W, V ) = Cov(3 + 6X, 20 7Y ) = 6( 7)(0.084) = 3.528. (7) Alternatively, since knowing the values of X and Y tells the corresponding values for W and V, we can use this information to construct the probability distribution table for W and V. Specifically, we have that: The probability distribution for W and V can be used directly to compute X = 0 W = 3 X = 1 W = 9 Y = 0 V = 20 0.15 0.07 Y = 1 V = 13 0.15 0.63 Cov(W, V ) using equation 2.24 in the book, which will give us Cov(W, V ) = 3.528. compute Cov(W, V ), the correlation results from using the equation: corr(w, V ) = However you Cov(W, V ) 3.528 = = 0.4425 (8) V ar(w )V ar(v ) (7.56)(8.4084) 2.6 This question considers the joint probability distribution between employment status and college graduation. a. The first question asks you to compute the mean of the employed random variable Y. This would be given by: b. The unemployment rate is given by E(Y ) = k Y p i y i = (0.05) (0) + (0.95) (1) = 0.95 Unemployment rate = Number employed Number in labor force = P r(y = 0) = 1 P r(y = 1) = 1 E(Y ) c. In order to compute the conditional means in this question, we need to know the conditional probabilities. These can be calculated using equation (2.17). For the current problem: P r(y = 0 X = 0) = P r(y = 1 X = 0) = P r(y = 0 X = 1) = P r(y = 1 X = 1) = P r(x = 0, Y = 0) P r(x = 0) P r(x = 0, Y = 1) P r(x = 0) P r(x = 1, Y = 0) P r(x = 1) P r(x = 1, Y = 1) P r(x = 1) = 0.045 0.754 = 0.0597 = 0.709 0.754 = 0.9403 = 0.005 0.246 = 0.0203 = 0.241 0.246 = 0.9797 3

The conditional means then follow using equation (2.18) in the text: E(Y X = 0) = (0) P r(y = 0 X = 0) + (1) P r(y = 1 X = 0) = 0.9403 E(Y X = 1) = (0) P r(y = 0 X = 1) + (1) P r(y = 1 X = 1) = 0.9797 d. The unemployment rate, conditional on the college graduation status, is computed in much the same way as the unemployment rate was computed in part (b), only now we are conditioning on graduation status. Thus, Similarly Unemployment rate X = 1 = Number employed among college graduates Number of college graduates in labor force = P r(y = 0 X = 1) = 1 P r(y = 1 X = 1) = 1 E(Y X = 1) = 0.0203 Unemployment rate X = 0 = Number employed among non-college graduates Number of non-college graduates in labor force = P r(y = 0 X = 0) = 1 P r(y = 1 X = 0) = 1 E(Y X = 0) = 0.0597 e. This question asks first what the probability of a randomly selected individual is unemployed. This is given by the P r[y = 0] = 0.05. You are then asked to compute conditional probabilities. That is, you are asked to compute the probability that an individual is a college graduate, given that you know they are unemployed. Mathematically, this corresponds to: P r(x = 1 Y = 0) = P r(x = 1, Y = 0) P r(y = 0) = 0.005 = 0.10 (9) 0.050 Thus, there is only a ten percent chance that a randomly selected unemployed person is a college graduate. The probability that a randomly selected unemployed person is a non-college graduate is then P r(x = 0 Y = 0) = 1 P r(x = 1 Y = 0) = 0.90 (10) f. In order to check independence, we need to determine if equation (2.22) in the text holds; i.e., However, we know that this does not hold, since P r(y = y X = x) = P r(y = y). (11) P r(y = 0 X = 0) = 0.0597 P r(y = 0) = 0.05. (12) 2.12 This question asks you to look up a number of the probabilities associated with specific distributions. a. The first question asks for P r(y > 1.75) if Y is distributed t 15. This corresponds to the 1-sided significance level in Table 2. Looking under the 15 degrees of freedom case, we find that the 1-sided significance level of 5% has a critical level of 1.75. That is, P r(y > 1.75) = 0.05, which is what we were looking for. b. The second question asks for P r( 1.99 Y 1.99) if Y is distributed t 90. However, this can be rewritten as: P r( 1.99 Y 1.99) = 1 [P r(y < 1.99) + P r(y > 1.99)] (13) 4

The term in square brackets above is just the 2-sided significance level in Table 2 (i.e., the probability that Y lies in the tails of the t-distribution - above 1.99 or below -1.99). Looking under the 90 degrees of freedom case, we find that the 2-sided significance level of 5% has a critical level of 1.99. That is, [P r(y < 1.99) + P r(y > 1.99)] = 0.05. This in turn implies that P r( 1.99 Y 1.99) = 1 [P r(y < 1.99) + P r(y > 1.99)] = 0.95 (14) c. The third question asks for P r( 1.99 Y 1.99) if Y is distributed N(0, 1). This can be rewritten as: P r( 1.99 Y 1.99) = P r(y 1.99) P r(y < 1.99) (15) From Table 1 we know that P r(y 1.99) = 0.9767 and P r(y < 1.99) = 0.0233, so that P r( 1.99 Y 1.99) = P r(y 1.99) P r(y < 1.99) = 0.9767 0.0233 = 0.9534. (16) d. The t-distribution and the standard normal are approximately the same as the degrees of freedom become large. e. The fifth question as for P r(y > 4.12) if Y is distributed F 4,7. This corresponds to the significance level. Looking through the three tables 5A, 5B and 5C, we see that for an F 4,7 distribution: P r(y > 2.96) = 0.10 P r(y > 4.12) = 0.05 P r(y > 7.85) = 0.01 The answer we are looking for is the second one; i.e., P r(y > 4.12) = 0.05. f. The sixth question as for P r(y > 2.79) if Y is distributed F 7,120. This corresponds to the significance level. Looking through the three tables 5A, 5B and 5C, we see that for an F 7,120 distribution: P r(y > 1.77) = 0.10 P r(y > 2.09) = 0.05 P r(y > 2.79) = 0.01 The answer we are looking for is the third one; i.e., P r(y > 2.79) = 0.01. 2.14 This question makes use of the central limit theorem, which says that as the sample size gets large, the sampling distribution for Ȳ can be approximated by N(µ Y, σ 2 ), where Ȳ This also means that Ȳ µ Ȳ σ Ȳ σ 2 Ȳ = σ2 Y n. (17) N(0, 1). (18) a. In this first question, you are told that n = 100, so that σ 2 Ȳ = σ2 Y n = 43 100 = 0.43. With this information, we know that: P r ( Ȳ 101 ) = P r ( ) Ȳ µ Ȳ 101 µ Ȳ (Ȳ ) µȳ 101 µȳ = P r σ Ȳ σ ( ) Ȳ 101 µȳ = Φ σ ( Ȳ ) 101 100 = Φ 0.43 Φ(1.525) = 0.9364. 5

b. In this second question, you are told that n = 165, so that σ 2 Ȳ = σ2 Y n = 43 165 information, we know that: P r ( Ȳ 98 ) = P r ( ) Ȳ µ Ȳ 98 µ Ȳ (Ȳ ) µȳ 98 µȳ = P r σ Ȳ σ ( ) Ȳ 98 µȳ = Φ σ ( Ȳ ) 98 100 = Φ 0.2606 =.2606. With this Φ( 3.92) 0.0000. We can then use this to calculate: P r ( Ȳ > 98 ) = 1 P r ( Ȳ 98 ) 1. (19) c. In this third question, you are told that n = 64, so that σ 2 Ȳ = σ2 Y n = 43 64 =.6719. With this information, we know that: P r ( 101 Ȳ 103) = P r ( 101 µ Ȳ Ȳ ) µȳ 103 µ Ȳ ( 101 µȳ = P r Ȳ ) µȳ 103 µȳ σ Ȳ σ Ȳ σ ( ) ( ) Ȳ 103 µȳ 101 µȳ = Φ Φ σ Ȳ σ ( ) ( Ȳ ) 103 100 101 100 = Φ Φ 0.6719 0.6719 Φ(3.66) Φ(1.22) 0.9999 0.8888 = 0.1111. 2.16 There are several ways to do this. Here is one way. Generate n draws of Y, Y 1, Y 2,... Y n. Let X i = 1 if Y i < 3.6, otherwise set X i = 0.. Notice that X i is a Bernoulli random variables with µ X = P r(x = 1) = P r(y < 3.6). Compute X. Because X converges in probability to µ X = P r(x = 1) = P r(y < 3.6), X will be an accurate approximation if n is large 6