Linear Regression Example Data

House Price in $1000s

(Y)

Square Feet
(X)

245

1400

312

1600

279

1700

308

1875

199

1100

219

1550

405

2350

324

2450

319

1425

255

1700
Statistics for Managers Using
Microsoft Excel, 5e 2008
Prentice-Hall, Inc.

Chap 13-1

Linear Regression Example

Scatterplot
House price model: scatter
plot

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-2

Linear Regression Example

Using Excel
Tools
-------Data Analysi
s
-------Regression

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-3

Linear Regression Example

Excel Output
The regression equation is:

Regression Statistics
Multiple R

0.76211

R Square

0.58082

Adjusted R Square

0.52842

Standard Error

house price 98.24833 0.10977 (square feet)

41.33032

Observations

10

ANOVA

Regression
Residual
Statistics
for Managers
Using Microsoft Excel, 5e
TotalPrentice-Hall,
2008
Inc.

df

SS

MS

18934.9348

18934.9348

13665.5652

1708.1957

Chap 13-4
32600.5000

Significance F

11.0848

0.01039

Linear Regression Example

Graphical Representation
House price model: scatter plot and regression line

Slope
= 0.10977

Intercept
= 98.248

house price 98.24833 0.10977 (square feet)

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-5

Linear Regression Example

Interpretation of b0
house price 98.24833 0.10977 (square feet)

b0 is the estimated mean value of Y whe

n the value of X is zero (if X = 0 is in the
range of observed X values)
Because the square footage of the hous
e cannot be 0, the Y intercept has no pr
actical application.
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-6

Linear Regression Example

Interpretation of b1
house price 98.24833 0.10977 (square feet)
b1 measures the mean change in the average v
alue of Y as a result of a one-unit change in X
Here, b1 = .10977 tells us that the mean value o
f a house increases by .10977($1000) = $109.77
, on average, for each additional one square fo
ot of size
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-7

Linear Regression Example

Making Predictions
Predict the price for a house with 2000 square feet:

house price 98.25 0.1098 (sq.ft.)

98.25 0.1098(200 0)
317.85
The predicted price for a house with 2000 square feet
is 317.85($1,000s) = $317,850
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-8

Linear Regression Example

Making Predictions
When using a regression model for prediction,
only predict within the relevant range of data
Relevant range for
interpolation

Do not try to
extrapolate beyond
the range of
observed Xs
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-9

Measures of Variation
Total variation is made up of two par
ts:

SST
Total Sum of
Squares

SSR

Regression Sum of
Squares

SST ( Yi Y )2

SSE

Error Sum of
Squares

)2
SSR ( Yi Y )2 SSE ( Yi Y
i

where:
=Y
Mean value of the dependent variable

Yi = Observed values of the dependent variable

Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Y
= Predicted value of Y for the given Xi value
Chap 13-10

Coefficient of Determination, r

The coefficient of determination is the porti

on of the total variation in the dependent va
riable that is explained by variation in the in
dependent variable
The coefficient of determination is also calle
d r-squared and is denoted as r2
SSR regression sum of squares
r

SST
total sum of squares
2

0 r 1
2

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-11

Linear Regression Example

Coefficient of Determination, r 2
SSR 18934.9348
r

0.58082
SST 32600.5000
2

Regression Statistics
Multiple R

0.76211

R Square

0.58082

Adjusted R Square

0.52842

Standard Error

58.08% of the variation in house

prices is explained by variation in
square feet

41.33032

Observations

10

ANOVA

df

SS

MS

Regression

18934.9348

18934.9348

Residual

13665.5652

1708.1957

32600.5000
Chap 13-12

Statistics for Managers

Total Microsoft Excel, 5e
Using
2008 Prentice-Hall,
Inc.

Significance F

11.0848

0.01039

Standard Error of Estimate

The standard deviation of the variation of observ
ations around the regression line is estimated by
n

SYX

SSE

n2

Where
SSE = error sum of squares
n = sample size
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-13

(
Y

Y
)
i i
i 1

n2

Linear Regression Example

Standard Error of Estimate
Regression Statistics
Multiple R

0.76211

R Square

0.58082

Adjusted R Square

0.52842

Standard Error

S YX 41.33032

41.33032

Observations

10

ANOVA

Regression
Residual for Managers
Statistics
Using Microsoft Excel, 5e
Total
2008 Prentice-Hall,
Inc.

df

SS

MS

18934.9348

18934.9348

13665.5652

1708.1957

Chap 13-14
32600.5000

Significance F

11.0848

0.01039

Comparing Standard Errors

SYX is a measure of the variation of
observed Y values from the regression
Y line
Y

small s YX

large s YX

The magnitude of SYX should always be judged relative to

the size of the Y values in the sample data
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-15

Inferences About the Slope:

t Test
t test for a population slope
Is there a linear relationship between X an
d Y?
Null and alternative hypotheses
H0: 1 = 0 (no linear relationship)
where:
does exist)
H1: 1 0 (linear relationship
b = regression slope
Test statistic
b1 1
coefficient

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Sb1

d.f.Chap
n13-16
2

1 = hypothesized slope
Sb1 = standard
error of the slope

Inferences About the Slope:

t Test Example
House Price in
$1000s
(y)

Square Feet
(x)

245

1400

312

1600

279

1700

308

1875

199

1100

219

1550

405

2350

324

2450

319

1425

255

1700

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Estimated Regression Equation:

house price 98.25 0.1098 (sq.ft.)

The slope of this model is 0.1098

Is there a relationship between the
square footage of the house and its
sales price?

Chap 13-17

Inferences About the Slope:

t Test Example
H 0 : 1 = 0

From Excel output:

H 1 : 1 0

Intercept
Square Feet

Coefficients

b1

Standard Error

Sb1
t Stat

98.24833

58.03348

1.69296

0.12892

0.10977

0.03297

3.32938

0.01039

b1 1 0.10977 0
t

3.32938
t
Sb1
0.03297
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

P-value

Chap 13-18

Inferences About the Slope:

t Test Example
Test Statistic: t = 3.329

H0: 1 = 0
H1: 1 0

d.f. = 10- 2
=8
/2=.02
5

Reject H0

Decision: Reject H0

/2=.025

Do not reject
H0
0

-t/2

2.3060

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Reject H

0
t/2
2.306 3.32
0
9

There is sufficient evidence

that square footage affects
house price

Chap 13-19

Inferences About the Slope:

t Test Example
P-Value

From Excel output:

Intercept

H0: 1 = 0
H1: 1 0

Square Feet

Coefficients

Standard Error

t Stat

98.24833

58.03348

1.69296

0.12892

0.10977

0.03297

3.32938

0.01039

Decision: Reject H0, since p-value <

There is sufficient evidence that
square footage affects house price.

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

P-value

Chap 13-20

F-Test for Significance

F Test statistic:
where

MSR
F
MSE
SSR
MSR
k
SSE
MSE
n k 1

where F follows an F distribution with k numerator degrees of freedom

and (n - k - 1) denominator degrees of freedom
(k = the number of independent variables in the regression model)

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-21

F-Test for Significance

Excel Output
Regression Statistics
Multiple R

0.76211

R Square

0.58082

Adjusted R Square

0.52842

Standard Error

41.33032

Observations

MSR 18934.9348
F

11.0848
MSE 1708.1957
With 1 and 8
degrees of
freedom

P-value
for the FTest

10

ANOVA

Regression
Statistics for Managers
Using
Microsoft Excel, 5e
Residual
2008 Prentice-Hall,
Inc.

df

SS

MS

18934.9348

18934.9348

13665.5652
Chap 13-22 1708.1957

F
11.0848

Significance F
0.01039

F-Test for Significance

Test Statistic:

H 0 : 1 = 0

MSR
F
11.08
MSE

H 1 : 1 0
= .05
df2 = 8
df1= 1

Decision:
Reject H0 at =
0.05
Conclusion:

Critical Value:
F = 5.32
=.
05

Do not
Reject H0
reject
Statistics
for
Managers
H0
F =
.05 5e
Using Microsoft Excel,
2008 Prentice-Hall,
Inc.

5.32

There is sufficient evidence that

house size affects selling price
Chap 13-23

Confidence Interval Estimate

for the Slope
Confidence Interval Estimate of the Slope:

b1 t n2Sb1

d.f. = n
-2

Excel Printout for House Prices:

Intercept
Square Feet

Coefficients

Standard Error

t Stat

P-value

Lower 95%

Upper 95%

98.24833

58.03348

1.69296

0.12892

-35.57720

232.07386

0.10977

0.03297

3.32938

0.01039

0.03374

0.18580

At the 95% level of confidence, the confidence

interval for the slope is (0.0337, 0.1858)
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-24

Confidence Interval Estimate

for the Slope
Coefficients

Intercept
Square Feet

Standard Error

t Stat

P-value

Lower 95%

Upper 95%

98.24833

58.03348

1.69296

0.12892

-35.57720

232.07386

0.10977

0.03297

3.32938

0.01039

0.03374

0.18580

Since the units of the house price variable is $1000s, you

are 95% confident that the mean change in sales price is
between $33.74 and $185.80 per square foot of house size
This 95% confidence interval does not include 0.
Conclusion: There is a significant relationship between house price and
square feet at the .05 level of significance

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-25

Estimating Mean Values and Predicting I

ndividual Values

Confidence
Interval for
the mean of
Y, given Xi

Goal: Form intervals around Y to express

uncertainty about the value of Y for a given Xi

Y = b0+b1Xi

Prediction Interval for an

individual Y, given Xi
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-26

Xi

Confidence Interval for

the Average Y, Given X
Confidence interval estimate for the mean value of
Y given a particular Xi

Confidence interval for Y|X Xi :

Y t n2S YX hi
Size of interval varies according to
distance away from mean, X

1 (X i X) 2 1
(X i X) 2
hi

n
SSX
n (X i X) 2

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-27

Prediction Interval for

an Individual Y, Given X
Prediction interval estimate for an individual value
of Y given a particular Xi

Prediction interval for YX X i :

Y t n 2SYX 1 hi

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

This extra term adds to the interval width to reflect the

added uncertainty for an individual case
Chap 13-28

Estimation of Mean Values: Example

Confidence Interval Estimate for Y|X=X

Find the 95% confidence interval for the mean price of 2,000
square-foot houses

Predicted Price Yi = 317.85

($1,000s)
2

1
(Xi X)

Y t n-2S YX

317.85 37.12
2
n (Xi X)

The confidence interval endpoints are 280.66 and

354.90,
or from $280,660 to $354,900
Statistics
for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-29

Estimation of Individual Values: Example

Prediction Interval Estimate for YX=X

Find the 95% prediction interval for an individual house with

2,000 square feet

Predicted Price Yi = 317.85 ($1,000s)

Y t n-1S YX

1
(Xi X)2
1
317.85 102.28
2
n (Xi X)

The prediction interval endpoints are 215.50 and 420.07, or

fromfor$215,500
to $420,070
Statistics
Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-30

t Test for a Correlation Coefficient

Hypotheses

H0 : = 0
H1 : 0

(no correlation between X and Y)

(correlation exists)

Test statistic
(with n 2rdegrees
of freedom)
-

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

where

1 r
n2
2

r r 2 if b1 0
r r 2 if b1 0
Chap 13-31

t Test for a Correlation Coefficient

Is there evidence of a linear relationship
between square feet and house price at the .05
level of significance?
H0: = 0 (No correlation)
H1: 0

(correlation exists)

=.05 , df = 10 - 2 = 8

t
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

r
1 r2
n2

.762 0
1 .762 2
10 2

Chap 13-32

3.329

t Test for a Correlation Coefficient

Decision:
Reject H0

d.f. = 10- 2 =
8
/2=.025

Reject H0

-t/2

2.3060
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

/2=.025

Do not reject
H
00

Conclusion:
There is evidence
of a linear
Reject H
t/2
association at the
2.3060
3.329 5% level of
significance
0

Chap 13-33

Residual Analysis
ei Yi Yi
The residual for observation i, e i, is the difference betwe
en its observed and predicted value
Check the assumptions of regression by examining the re
siduals
Examine for Linearity assumption
Evaluate Independence assumption
Evaluate Normal distribution assumption
Examine Equal variance for all levels of X
Statistics
Graphical
Analysis of Residuals
for Managers
Using Microsoft Excel, 5e
Chap 13-34
Can
plot residuals vs. X
2008
Prentice-Hall,
Inc.

Residual Analysis for Linearity

Y

x
Not
Linear

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

residuals

residuals

x
Linear

Chap 13-35

Residual Analysis for Independ

ence

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

residuals

residuals

Independent

residuals

Not Independent

Chap 13-36

Checking for Normality

Examine the Stem-and-Leaf Display of th
e Residuals
Examine the Box-and-Whisker Plot of th
e Residuals
Examine the Histogram of the Residuals
Construct a Normal Probability Plot of th
e Residuals
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-37

Residual Analysis for

Equal Variance
Y

residuals

residuals

Unequal variance

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

x
Equal variance

Chap 13-38

Linear Regression Example

Excel Residual Output
RESIDUAL OUTPUT
Predicted
House Price

Residuals

251.92316

-6.923162

273.87671

38.12329

284.85348

-5.853484

304.06284

3.937162

218.99284

-19.99284

268.38832

-49.38832

356.20251

48.79749

367.17929

-43.17929

254.6674

64.33264

10

284.85348

-29.85348

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Does not appear to violate

any regression assumptions

Chap 13-39

Measuring Autocorrelation:
The Durbin-Watson Statistic
Used when data are collected over time
to detect if autocorrelation is present
Autocorrelation exists if residuals in on
e time period are related to residuals in
another period

Statistics for Managers

Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-40

Autocorrelation
Autocorrelation is correlation of the errors (r
esiduals) over time

Here, residuals suggest a

cyclic pattern, not
random

Violates the regression assumption that residuals are

statistically independent
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-41

Strategies for Avoiding

the Pitfalls of Regression
Start with a scatter plot of X on Y to obser
ve possible relationship
Perform residual analysis to check the ass
umptions
Plot the residuals vs. X to check for violations
of assumptions such as equal variance
Use a histogram, stem-and-leaf display, box-a
nd-whisker plot, or normal probability plot of
the residuals to uncover possible non-normal
Statistics ity
for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-42

Strategies for Avoiding

the Pitfalls of Regression
If there is violation of any assumption, us
e alternative methods or models
If there is no evidence of assumption viol
ation, then test for the significance of the
regression coefficients and construct conf
idence intervals and prediction intervals
Avoid making predictions or forecasts out
side the relevant range
Statistics for Managers
Using Microsoft Excel, 5e
2008 Prentice-Hall,
Inc.

Chap 13-43