Scribd
Upload
172 views4 pages

UCCM2263 Tutorial 1 2016

This document contains 7 examples demonstrating concepts and calculations for simple linear regression models. The examples cover: 1) calculating least squares estimates of slope and intercept parameters, 2) computing t-statistics and confidence intervals, 3) deriving the covariance between parameter estimates, 4) calculating mean squared error, 5) constructing prediction intervals, 6) fitting regression models to medical data and testing hypotheses, and 7) additional practice problems involving properties of the least squares regression line and confidence intervals.

Uploaded by

sadyehclen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
172 views4 pages

UCCM2263 Tutorial 1 2016

This document contains 7 examples demonstrating concepts and calculations for simple linear regression models. The examples cover: 1) calculating least squares estimates of slope and intercept parameters, 2) computing t-statistics and confidence intervals, 3) deriving the covariance between parameter estimates, 4) calculating mean squared error, 5) constructing prediction intervals, 6) fitting regression models to medical data and testing hypotheses, and 7) additional practice problems involving properties of the least squares regression line and confidence intervals.

Uploaded by

sadyehclen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Chapter 1: Linear Regression Model

Tutorial 1
1.

A sample of 10 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:
10

10

10

10

10

i =1

i =1

i =1

i =1

i =1

xi =100; y i = 200;

2
2
xi y i = 2000; xi = 2000; y i = 5000

Calculate the least squares estimate of 0 and 1 .


2.

A sample of 10 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:
10

10

i =1

i =1

2
2
( xi x ) = 400; ( y i y ) = 425;

10

2
( y i y ) = 225

i =1

Calculate the t-statistic used for testing the hypothesis H 0 : 1 = 0 .


3.

a)

Derive Cov( 0 , 1 ) .

b)

A sample of 20 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:
20

20

i =1

i =1

2
xi =200; xi = 2400; MS E = 100. Determine the estimated covariance between 0 and 1 .

4.

A sample of 10 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:
10

10

i =1

i =1

y i = 30;

5.

2
2
y i = 690; R = .6. Calculate MS E .

A sample of 8 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:

1 = 35.69 ; ( xi x ) 2 =1.62; ( y i y ) 2 = 2394


Determine the 90% confidence interval for 1 .

UCCM2263/UECM2263 Applied Statistical Models

Chapter 1: Linear Regression Model


6.

A sample of 20 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .
The summaries of the data are:

0 = 138.561; 1 = 1 .104; ( xi x ) 2 = 10.668 ; ( y i y ) 2 = 20.838;


2
2
ei = ( y i y ) = 7.832; x = 2

a)
b)

7.

Construct a 99% confidence interval for the mean per capita consumption of natural gas ( E (Yh ) )
for all gas stations with the prices of natural gas at $2.11.
Construct a 99% prediction interval for the per capita consumption of natural gas Yh for a gas
stations with the prices of natural gas at $2.11.

In a medical survey, the weight X (in pounds) and systolic blood pressure Y of 26 randomly selected
males in the age group 25 30 are recorded and the summaries are given below.
26

xi = 4743;
i =1
26

(x
i =1

(a)
(b)
(c)

26

yi = 3786;
i =1

x ) 2 = 15312.35;

26

xi yi = 697076;
i =1

26

( y

26

xi2 = 880545;
i =1

y ) 2 = 4502.15;

i =1

26

( y

26

2
i

= 555802;

i =1

y i ) 2 = 1808.5726

i =1

(i)
Fit a simple regression model to the data.
(ii)
Find a 99% confidence interval on the slope. Interpret the interval obtained.
Construct the ANOVA table and test for significance of regression by using the significance
level, = 0.05 .
(i)
Construct a 95% confidence interval on the mean response with x h =180 lb . Interpret the
interval obtained.
(ii)
Find the coefficient of determination and explain what it means.

UCCM2263/UECM2263 Applied Statistical Models

Chapter 1: Linear Regression Model


Take home exercise:
1.

A sample of 20 pairs of values ( xi , y i ) which will be represented by the following model:

yi = 0 + 1 xi + i .

F ratio for testing the hypothesis H 0 : 1 = 0 is equal to 12. Calculate R 2 .


2.

The number of pounds of steam used per month at a plant is thought to be related to the average monthly
ambient temperature. Last years usage Y (in 1000 pounds) and temperatures X (in o F ) were studied,
the summaries of the data is given below.
12

12

i =1

i =1

12

12

12

i =1

i =1

i =1

2
2
xi = 558; y i = 5062.34; xi y i = 265869.63; xi = 29256; y i = 2416234.61;
12

12

12

i =1

i =1

i =1

2
2
( xi x ) = 3309; ( y i y ) = 280627.4204;

(a)

(i)

2
( y i y i ) = 37.8547.

Fit a simple regression model to the data. State one of the assumptions made in this
model. Interpret the meaning of

found.

State the mean and variance for the least square estimator of 1 .
State the relationship between the correlation coefficient and the slope, and hence find the
correlation coefficient and interpret its meaning.
Plant management believes that an increase in average ambient temperature of 1 o F will increase
average monthly steam usage by 9100 lb . Do the data support this statement, use the
significance level, = 0.01 ?
Construct a 99% prediction interval on y h with x h = 58 o F . Interpret the interval obtained.

(ii)
(iii)
(b)

(c)

1) Consider the simple linear regression model y = 0 + 1 x + , with E ( ) = 0 , Var ( ) = 2 and


uncorrelated.
2

a) Show that Cov 0 , 1 = x
.
S xx


b) Show that Cov y, 1 = 0 .

2) Show that the sum of residual equals zero,

= 0.

i =1

3) An alternative model of the simple linear regression is y i = 0* + 1 xi x + i where

*
S xy
remains
0* = 0 + 1 x . The least squares estimator for the parameters are 0 = y and 1 =
S xx
the same as before. Show that the least squares regression line always passes through the centroid
x, y of the data.

( )

UCCM2263/UECM2263 Applied Statistical Models

Chapter 1: Linear Regression Model


4) Consider the simple linear regression model y = 0 + 1 x + where the intercept 0 is known.
a) Find the least squares estimator of 1 for this model. Does this answer seem reasonable?


b) What is the variance of the slope 1 for the least squares estimator found in part (a)?

c) Find a 100(1 )% confidence interval for 1 . Is this interval narrower than the estimator for
the case where both slope and intercept are unknown?
5) Show that the coefficient of determination has property 0 R 2 1 .

UCCM2263/UECM2263 Applied Statistical Models

You might also like