MATH 423 midterm -2007

# MATH 423 midterm -2007 - Student Name: Student Number:...

This preview shows pages 1–4. Sign up to view the full content.

Student Name: Student Number: Faculty of Science MIDTERM EXAMINATION Mathematics 423 Regression and Analysis of Variance Sample Midterm You must do all THREE questions. Calculators are allowed. One 8.5” × 11” sheet of notes is allowed. Language dictionaries are allowed. There are 8 pages to this exam and the total number of marks for the exam is 100. 1

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

View Full Document
Question 1: Suppose that in the standard linear regression model that (1) y i = β 0 + β 1 x i + ǫ i where ǫ i Normal(0 , γ 2 i σ 2 ) (i.e. are normally distributed with mean 0 and variance γ 2 i σ 2 ) where the γ i values are all known constants, but σ 2 is unknown. (a) List the usual assumptions of the Simple Linear Regression model. (b) Can we use the usual Simple Linear Regression model to estimate the parameters β 0 and β 1 ? Why or why not? Now consider the following regression model: (2) γ - 1 i y i = γ - 1 i β 0 + γ - 1 i β 1 x i + γ - 1 i ǫ i (c) Show that model 2 satisFes the assumptions of the standard statistical model. Hint: Re- write model (2) as z i = u i β 0 + v i β 1 + δ i and relate z i , u i , v i and δ i to the variables in the original model. (d) ±ind the least squares estimates of β 0 and β 1 . (e) Show that performing least squares on the new model in part (b) is equivalent minimizing n i =1 ( y i - β 0 + β 1 x i ) 2 ( γ i ) - 2 . 2
Consider the usual multiple linear regression model, Y = X β + ǫ. (a) Show (via matrix algebra) that the sample residuals are orthogonal to every column of X . Hint: Write the residuals in terms of Y , X , and ˆ β . (b) Show that the sum of the sample residuals is equal to zero. Hint: Use part (a). Sometimes, if the dimension of X is large, it is computationally easier to work with the Q-R decomposition of X , where X = Q R where Q is an orthogonal matrix ( Q t Q = I ) and R is upper-triangular ( r ij = 0 for i > j ). This is because, with a Q-R decomposition, we do not ever have to calculate the matrix X t X . (c) Show that

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 01/15/2010 for the course MATH 423 taught by Professor Steele during the Spring '06 term at McGill.

### Page1 / 8

MATH 423 midterm -2007 - Student Name: Student Number:...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online