Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
44 views

Regression

This document contains sample questions related to simple linear regression, multiple linear regression, and their applications. The questions cover topics such as the coefficient of determination, residual vectors, rank of the design matrix, deleted residuals, maximum likelihood estimation of coefficients, covariance of estimated coefficients, hat matrix, and interpretation of regression coefficients.

Uploaded by

Luyanda Blom
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
44 views

Regression

This document contains sample questions related to simple linear regression, multiple linear regression, and their applications. The questions cover topics such as the coefficient of determination, residual vectors, rank of the design matrix, deleted residuals, maximum likelihood estimation of coefficients, covariance of estimated coefficients, hat matrix, and interpretation of regression coefficients.

Uploaded by

Luyanda Blom
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Sample questions

Simple linear regression

1. Given a simple linear regression model yi = β0 +β1 xi +εi , i=1,2. . . ,n.


With E (εi ) = 0, V (εi ) = σ 2 , and ε0i s uncorrelated

a) Show that for this model the coefficient of determination (R2 ) is


equivalent to the r2 (the square of the correlation coefficient)
b) Show that the residual vector (ei ) is correlated with the vector
of observed yi

Multiple linear regression

2. For a multiple linear regression model, Y = Xβ + ε, E (ε) = 0, and


V (ε) = σ 2 I

a) Given that the rank of X is such that p + 1 < n, where p, is the


number of predictors and n, the number of observations. Prove
0
that X X is positive definite
ei
b) Show that the deleted residual (e(i) ) can be expressed as
1 − hii
c) Use the maximum likelihood method to obtain the estimate of β
d) Derive the covariance matrix of β̂ and hence state its distribution
e) Show that HY = Ŷ , where H is the hat matrix

Application questions

1 a) A roket motor is manufactured by joining an igniter propellant and


a sustainer propellant together inside a metal housing. The Shear
strength of the bond between the two types of propellant is an important
characteristic. Looking at FIGURE 1 below, would you recommend
a simple linear regression model for shear strength(y) and propellant
age (x)?

b) Calculate the standard error for βˆ0 and βˆ1 using values stated in
TABLE 1 from an output of fitting a simple linear regression model
on the dataset described above and use it to calculate the confidence
intervals for the estimators at α = 0.05.

1
2

Scatter plot

2500

2250
Shear Strength

2000

1750

5 10 15 20 25
Propellant Age

Figure 1

c) What does the model suggest about the relationship between shear
strength and the age of the propellant.

d) State the model ŷ and find a 95% predictive interval on a future value
of x = 10 weeks.

2) The output of a linear regression model fitted on two predictor variable


is presented in TABLE 2. Note that the second predictor variable is
an indicator variable, such that, x2 = 0, x2 = 1 and therefore βˆ3 is
the coefficient of the interaction (x1 ∗ x2 ) . Using TABLE 2, state
3

Table 1

Coefficients βˆ0 := 2627.822 βˆ1 :=-37.154


t value 59.48 -12.86
Residual standard error 96.11
Multiple R-squared 0.9018
Adjusted R-squared 0.8964
F-statistic 165.4
n 20
x̄ 13.3625
Sxx 1106.559

Table 2

Coefficients βˆ0 := 33.8383695 βˆ1 :=-0.1015306 βˆ2 :=8.131250 βˆ3 =-0.0004171


t value 13.864 −7.779 2.225 −0.023
Residual standard error 3.32
Multiple R-squared 0.8951
Adjusted R-squared 0.8754
F-statistic 45.49
n 20

the final model for each of the groups. Justify your answer at a 95%
confidence level.

You might also like