Lecture_3 - Linear Regression with One Regressor Linear...

Info iconThis preview shows pages 1–13. Sign up to view the full content.

View Full Document Right Arrow Icon
1 Linear Regression with One Regressor Linear regression allows us to estimate, and make inferences about, population slope coefficients. Ultimately our aim is to estimate the causal effect on Y of a unit change in X
Background image of page 1

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

View Full Document Right Arrow Icon
2 Estimation: How should we draw a line through the sample data to estimate the (population) slope (answer: ordinary least squares). Inference: Testing: How to test if the slope is zero? Confidence intervals for the slope The problems of statistical inference for linear regression are similar to estimation of the mean
Background image of page 2
3 Linear Regression: Some Notation and Terminology The population regression line or the (linear) conditional expectation function : E(Avg Test Score|STR) = 0 + 1 STR 1 = slope of population regression line = avg. chg. in test score for unit change in STR Why are 0 and 1 “population” parameters ? We would like to know the population value of 1 . We don‟t know 1 , so must estimate it using data.
Background image of page 3

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

View Full Document Right Arrow Icon
Conditional Expectation 4 Is the function of X that best predicts Y Recall that Y was the best predictor of Y in population: it solves. Now instead of predicting Y by a constant, we use any function of X. The function that does the trick: E(Y|X=x) We will make the assumption that this is linear in a given set of X‟s. This will be our OLS assumption 1
Background image of page 4
5 The Population Linear Regression Model – general notation Y i = 0 + 1 X i + u i , i = 1,…, n X is the independent variable or regressor Y is the dependent variable 0 = intercept 1 = slope u i = the regression error, note by defn. of CEF E(u|X) = 0 The regression error consists of omitted factors, or possibly measurement error in the measurement of Y . In general, these omitted factors are other factors that influence Y
Background image of page 5

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

View Full Document Right Arrow Icon
6 This terminology in a picture : Observations on Y and X ; the population regression line; and the regression error (the “error term”):
Background image of page 6
7 The Ordinary Least Squares Estimator How can we estimate 0 and 1 from data? Analogy principle We will focus on the least squares (“ ordinary least squares ” or OLS ”) estimator of the unknown parameters 0 and 1 , which solves the same problem as the (linear) CEF in the sample: 01 2 , 0 1 1 min [ ( )] n b b i i i Y b b X
Background image of page 7

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

View Full Document Right Arrow Icon
8 Mechanics of OLS The population regression line: Test Score = 0 + 1 STR 1 = Test score STR = ??
Background image of page 8
9 The OLS estimator solves: 01 2 , 0 1 1 min [ ( )] n b b i i i Y b b X The OLS estimator minimizes the average squared difference between the actual values of Y i and the prediction (“predicted value”) based on the estimated line. This minimization problem can be solved using calculus (App. 4.2). The result is the OLS estimators of 0 and 1 .
Background image of page 9

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

View Full Document Right Arrow Icon
10
Background image of page 10
11 Two properties of OLS Sample mean of residuals is zero (mimics E(u) = 0 in population) Sample covariance of residuals and predicted values is 0 (mimics the population equation: cov(X, u) = 0)
Background image of page 11

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

View Full Document Right Arrow Icon
12 Application to the California Test Score – Class Size data Estimated slope = 1 ˆ = – 2.28 Estimated intercept = 0 ˆ = 698.9 Estimated regression line: TestScore = 698.9 – 2.28 STR
Background image of page 12
Image of page 13
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 34

Lecture_3 - Linear Regression with One Regressor Linear...

This preview shows document pages 1 - 13. Sign up to view the full document.

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