Class 13,14 June 17, 19, 2015

Similar documents
Econometric Methods. Review of Estimation

Lecture 3. Sampling, sampling distributions, and parameter estimation

{ }{ ( )} (, ) = ( ) ( ) ( ) Chapter 14 Exercises in Sampling Theory. Exercise 1 (Simple random sampling): Solution:

Chapter 8: Statistical Analysis of Simulated Data

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS

hp calculators HP 30S Statistics Averages and Standard Deviations Average and Standard Deviation Practice Finding Averages and Standard Deviations

Discrete Mathematics and Probability Theory Fall 2016 Seshia and Walrand DIS 10b

Continuous Distributions

Parameter, Statistic and Random Samples

Chapter 5 Properties of a Random Sample

Lecture 7. Confidence Intervals and Hypothesis Tests in the Simple CLR Model

UNIVERSITY OF OSLO DEPARTMENT OF ECONOMICS

Chapter 3 Sampling For Proportions and Percentages

Special Instructions / Useful Data

ENGI 4421 Joint Probability Distributions Page Joint Probability Distributions [Navidi sections 2.5 and 2.6; Devore sections

THE ROYAL STATISTICAL SOCIETY 2016 EXAMINATIONS SOLUTIONS HIGHER CERTIFICATE MODULE 5

Simulation Output Analysis

Law of Large Numbers

Summary of the lecture in Biostatistics

GOALS The Samples Why Sample the Population? What is a Probability Sample? Four Most Commonly Used Probability Sampling Methods

X ε ) = 0, or equivalently, lim

Chapter 8. Inferences about More Than Two Population Central Values

Lecture Notes Types of economic variables

STATISTICAL PROPERTIES OF LEAST SQUARES ESTIMATORS. x, where. = y - ˆ " 1

X X X E[ ] E X E X. is the ()m n where the ( i,)th. j element is the mean of the ( i,)th., then

CHAPTER VI Statistical Analysis of Experimental Data

Qualifying Exam Statistical Theory Problem Solutions August 2005

best estimate (mean) for X uncertainty or error in the measurement (systematic, random or statistical) best

Mean is only appropriate for interval or ratio scales, not ordinal or nominal.

STA 105-M BASIC STATISTICS (This is a multiple choice paper.)

PROPERTIES OF GOOD ESTIMATORS

8.1 Hashing Algorithms

Chapter -2 Simple Random Sampling

STATISTICAL INFERENCE

CS286.2 Lecture 4: Dinur s Proof of the PCP Theorem

To use adaptive cluster sampling we must first make some definitions of the sampling universe:

Point Estimation: definition of estimators

Chapter -2 Simple Random Sampling

Lecture 1 Review of Fundamental Statistical Concepts

22 Nonparametric Methods.

Lecture 4 Sep 9, 2015

ENGI 3423 Simple Linear Regression Page 12-01

THE ROYAL STATISTICAL SOCIETY GRADUATE DIPLOMA

( ) = ( ) ( ) Chapter 13 Asymptotic Theory and Stochastic Regressors. Stochastic regressors model

f f... f 1 n n (ii) Median : It is the value of the middle-most observation(s).

Investigation of Partially Conditional RP Model with Response Error. Ed Stanek

Statistics Descriptive and Inferential Statistics. Instructor: Daisuke Nagakura

MEASURES OF DISPERSION

Module 7: Probability and Statistics

Measures of Dispersion

Third handout: On the Gini Index

Midterm Exam 1, section 2 (Solution) Thursday, February hour, 15 minutes

Random Variables and Probability Distributions

THE ROYAL STATISTICAL SOCIETY HIGHER CERTIFICATE

1. The weight of six Golden Retrievers is 66, 61, 70, 67, 92 and 66 pounds. The weight of six Labrador Retrievers is 54, 60, 72, 78, 84 and 67.

2.28 The Wall Street Journal is probably referring to the average number of cubes used per glass measured for some population that they have chosen.

Lecture 3 Probability review (cont d)

STK4011 and STK9011 Autumn 2016

THE ROYAL STATISTICAL SOCIETY 2016 EXAMINATIONS SOLUTIONS GRADUATE DIPLOMA MODULE 2

ENGI 4421 Propagation of Error Page 8-01

BIOREPS Problem Set #11 The Evolution of DNA Strands

Homework 1: Solutions Sid Banerjee Problem 1: (Practice with Asymptotic Notation) ORIE 4520: Stochastics at Scale Fall 2015

Chapter 11 Systematic Sampling

Professor Wei Zhu. 1. Sampling from the Normal Population

Simple Linear Regression

Multiple Linear Regression Analysis

9 U-STATISTICS. Eh =(m!) 1 Eh(X (1),..., X (m ) ) i.i.d

Midterm Exam 1, section 1 (Solution) Thursday, February hour, 15 minutes

Module 7. Lecture 7: Statistical parameter estimation

Example: Multiple linear regression. Least squares regression. Repetition: Simple linear regression. Tron Anders Moger

Objectives of Multiple Regression

ESS Line Fitting

Introduction to local (nonparametric) density estimation. methods

Lecture Notes to Rice Chapter 5

LECTURE - 4 SIMPLE RANDOM SAMPLING DR. SHALABH DEPARTMENT OF MATHEMATICS AND STATISTICS INDIAN INSTITUTE OF TECHNOLOGY KANPUR

Chapter 4: Elements of Statistics

Ordinary Least Squares Regression. Simple Regression. Algebra and Assumptions.

A Study of the Reproducibility of Measurements with HUR Leg Extension/Curl Research Line

The Occupancy and Coupon Collector problems

Chapter 4 Multiple Random Variables

12.2 Estimating Model parameters Assumptions: ox and y are related according to the simple linear regression model

Simple Linear Regression

Section 2 Notes. Elizabeth Stone and Charles Wang. January 15, Expectation and Conditional Expectation of a Random Variable.

Chapter 9 Jordan Block Matrices

Uncertainty, Data, and Judgment

The expected value of a sum of random variables,, is the sum of the expected values:

CHAPTER 3 POSTERIOR DISTRIBUTIONS

Bayes Interval Estimation for binomial proportion and difference of two binomial proportions with Simulation Study

Laboratory I.10 It All Adds Up

2SLS Estimates ECON In this case, begin with the assumption that E[ i

Bayes (Naïve or not) Classifiers: Generative Approach

Lecture 2 - What are component and system reliability and how it can be improved?

Chapter 4 Multiple Random Variables

å 1 13 Practice Final Examination Solutions - = CS109 Dec 5, 2018

Dr. Shalabh. Indian Institute of Technology Kanpur

Analysis of Variance with Weibull Data

ρ < 1 be five real numbers. The

Sampling Theory MODULE X LECTURE - 35 TWO STAGE SAMPLING (SUB SAMPLING)

1 Solution to Problem 6.40

Multiple Choice Test. Chapter Adequacy of Models for Regression

Transcription:

Class 3,4 Jue 7, 9, 05 Pla for Class3,4:. Samplg dstrbuto of sample mea. The Cetral Lmt Theorem (CLT). Cofdece terval for ukow mea.. Samplg Dstrbuto for Sample mea. Methods used are based o CLT ( Cetral Lmt Theorem) ad rules of expectato ad varace. O class 6 we troduced a radom varable havg meag of a w a card game. was defed as follows: If a card draw s er Jack or Quee, (s) = $5 ; f s s er Kg or Ace, (s)=0$. Orwse (s)=-4. Dstrbuto of : -4 5 0 p 36/5 8/5 8/5

I frst row possble values are gve, secod row correspodg probabltes. The Expectato (Expected value, Mea) s defed as follows: E( ) p x ; Expectato also s deoted by E( ). I our problem of drawg a card 8 8 E( ) px 5 0 5 5 The Varace V ( ) s p ( x defed ) as p x 36 4 5. 0.465. The last expresso (rght-had-sde) s so called a shortcut formula for varace. Usg t we fd V ( ) V ( ) p x 6 30.4, ( ) 5.5. 36 5 5 8 5 00 8 5 0.465 30.3.

Example. Assume we are playg ow games depedetly,, deote r ws respectvely. Now cosder sum S Soluto. of S has two games. followg Fd P( S ). dstrbuto : S -8 + 6 0 5 0 p 0.479 0.3 0.3 0.04 0.048 0.04 The resultg table s called Samplg Dstrbuto of Sum. The table gves us P ( S ) 0.479. Example. Fd P ( ). More frequetly dstrbuto of followg close relatve of S s cosdered: ( + ) / whch s called Samplg Dstrbuto of Sample Mea ( + )/ -4 +/ 3 5 7.5 0 p 0.479 0.3 0.3 0.04 0.048 0.04 So, soluto to Example s: P ( ) 0.69.

Example 3. Now we cosder 00 games ad put k { The w k th game}, k,...,00, ad form sample mea :... ( our case equals 00). Fd P( ). Now ma questo s: What s dstrbuto of, or speakg statstcal laguage, what s samplg dstrbuto of sample mea? CLT helps solve out ths problem. A smplfed verso of CLT whch s frequetly used appled statstcs souds as follows: The sum of may depedet detcally dstrbuted radom varables havg a varace, has approxmately a Normal dstrbuto. So CLT tells us that samplg dstrbuto of s approxmately Normal. It remas to fd mea ad varace of. The formulas for ad ; are :. () We ca easly prove above formulas by use of followg rules for expectatos ad varaces:

E( a V ( V ( a )... a... ) a V ( ), ad ) V ( a E( for )... a depedet )... V ( ). E( ); (4) (3)... Remark. We ote that (3) holds wthout codto of depedece, whle (4) depedece s eeded. We also remark (warg!) (4) s ot correct for stadard devatos. Now we come back to our Example 3. Accordg to above facts, 5.5 ~ N ( 0.465,( ) ) 00, ad we fd 4.65 P( ) P( Z ) P( Z.66) 0.0039. 5.5. Homework Assgmet. Exercses 5.6-5., p.83; 5.4

Reducto of arbtrary Normal probablty to that of Stadard Normal. Assume P( a ~ N(, ). The a b b) P( Z ). Example 4. Assume we select at radom a battery from a large batch of alkale batteres. The radom varable of terest s = {The lfetme of a radomly chose battery} Assume t s kow that 0 hours ad hours. Now choose depedetly at radom 00 batteres, cosder 00 depedet radom varables: k { The lfetme of k th battery }, k,...,00, ad form sample mea :... ( our case equals 00). What s P( 9.8)?

Accordg to CLT, has approxmately a Normal dstrbuto. I order to fd desred probablty we have to calculate ad : 0; 0.. () ~ N (0,(0.) ), ad we fd P ( 9.8) 0. 6. Aor sub-problem o your tests mght be: Fd such that ( 9.8) 0.05. P (Aswer: =7). Soluto. P( 9.8) 0.05. Ths mples 9.8 0 /.645. Ths mples P( Z 7. 9.8 0 ) / 0.05. By table, Homework Assgmet o Samplg Dstrbuto of Sample Mea: Exercses 5.6-5., p.83; 5.4.

Cofdece Iterval for ukow mea. Exercse. Cosder followg radom varable ={The mleage per gallo for a radomly chose car of a ew model}, s ot kow. such kd of problem has a very clear practcal meag: t s a populato (global) average of mleage per gallo. The problem s stated as follows: Gve a sample,..., for radomly chose =00 cars of model fd a 95%-cofdece terval for x 3 mles / gallo, 4.6 mles / gallo. The values are gve.

A has For ( ) 00% cofdece cofdece form : z /, where s level Most Come of umber for frequetly level () whch 95% t 0.05, ad back 00, cofdece to sze our 4.6 of sample, ( ) 00% s z / P( z practcal.96. t erval / z cosdered. / ad Z s statstca l example. x 3, z / z for depeds /.96, Ad resultg cofdece terval s 3 0. 90. 0.90 s called boud for error ukow defed ) o. problems. as

Fdg sample sze. Assume we wat to have boud for error 0.50 (stead of 0.90). What s sample sze that esures ths boud for error at same level of cofdece? The problem reduces to followg equato wth respect to : 4.6.96 0.5; 36. Homework Assgmets: Exercses: 6.-6.5, p.307.