Ch. 4-5 Linear Regression with One Regressor

Ch. 4-5 Linear Regression with One Regressor - For...

Info iconThis preview shows pages 1–11. Sign up to view the full content.

View Full Document Right Arrow Icon
1 | One Regressor For Probability and Statistics, many more exercise questions are available from the textbook. The exercise questions in the textbook that you are not responsible for: Chapter 2: (2.16), (2.19) – (2.21), (2.23); and (2.25) – (2.27) from the 3 rd ed. Chapter 3: (3.2) – (3.5), (3.7), (3.9), (3,10), (3.12), (3.13), (3.14), (3,15), (3.16) c,d ; (3.17) – (3.21) .
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 | One Regressor 3. LINEAR REGRESSION WITH ONE REGRESSOR (CH. 4 - 5) The exercise questions in the textbook that you are not responsible for: Chapter 4: (4.4), (4.8), (4.10); and (4.13), (4.14) from the 3 rd ed. Chapter 5: (5.6), (5.9), (5.11), (5.12), (5.15).
Background image of page 2
3 | One Regressor [1] Motivation • Wish to estimate the average relation between class size and students’ performances. • Set up the relation by the following equation: 0 ClassSize TestScore ClassSize other factors  , (1) where TestScore means standardized test score and ClassSize means student-teacher ratio. ClassSize measures the average change in TestScore when ClassSize changes by one unit. We wish to estimate ClassSize .
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4 | One Regressor [2] Regression Model with a Single Regressor (1) General linear regression model with a single regressor 01 YX u  , (2) where: Y is the dependent variable (or regressand , or left-hand variable ); X is the independent variable (or regressor, right-hand variable ); β 0 + β 1 X is the population regression line ( function ); β 0 is the intercept of the population regression line; β 1 is the slope (or coefficient ) on the population regression line; and u is the error term .
Background image of page 4
5 | One Regressor (2) Least Squares Assumptions: Assumption 1: Conditional mean of u given X is zero; E ( u | X ) = 0. u cannot be explained by X . No correlation between u and X . 01 ( | )( | ) E YX E X uX X   . 1 () o X describes the average relation between Y and X . Assumption 2: X and u have finite-fourth moments. E ( x 4 ), E ( u 4 ) and E ( x 2 u 2 ) exist and are finite (< ): Technical assumption for CLT. Generally holds.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
6 | One Regressor Assumption 3: A random sample ( X i , Y i ), i = 1, 2, . .. , n is available. To obtain unbiased and consistent estimates of β 0 and β 1 . Likely to be violated if time-series data are used. The estimators of β 0 and β 1 we will learn could be consistent even if this is violated.
Background image of page 6
7 | One Regressor
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
8 | One Regressor • What can we do with regression analysis? Y = 0 + 1 X + u , where, Y = TestScore and X = ClassSize. Or, Y = teaching evaluation and X = teacher’s beauty. Or, Y = wage and X = beauty. Or, Y = GDP (per capita) growth and X = past per capital GDP.
Background image of page 8
9 | One Regressor Estimation : Find good approximates of β 1 and β 2 (that is, estimate an average relation between Y & X ). Hypothesis Testing : • A taxpayer claims that ClassSize has no effect on TestScore. • Beauty does not influence teaching evaluation (See the data set named “TeachingRatings”) Forecasting and Prediction : Wish to predict the average test score of the district with average student-teacher ratio = 14.
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
10 | One Regressor [3] Ordinary Least Squares (OLS) Estimation - Gauss Let {( X i , Y i )| i = 1, . .. , n } be a random sample.
Background image of page 10
Image of page 11
This is the end of the preview. Sign up to access the rest of the document.

This document was uploaded on 02/10/2012.

Page1 / 56

Ch. 4-5 Linear Regression with One Regressor - For...

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online