LECTURE 12

Simple Linear Regression

 Learning Objectives
In this lecture, you learn:
 How to calculate and test a correlation coefficient
 How to use regression analysis to predict the value
of a dependent variable
 To compute and interpret regression coefficients
 To compute and interpret coefficient of
determination
 To inference about the slope
 To estimate mean values and individual values
 Correlation coefficient
 The formula used to compute the linear correlation
coefficient:
 Correlation coefficient
 To simplify the notation, we define three terms called
sums of squares

 The formula for the sample correlation coefficient can

be re-written:
Correlation coefficient and
scatter plots
 Test for significant correlation
 Step 1: State the Hypotheses
 H0:  = 0 (population correlation coefficient equals zero)
 H1:  ≠ 0
 Step 2: Specify the Decision Rule
 For degrees of freedom df = n -2, look up the critical value t .
 Step 3: Calculate the Test Statistic

 Step 4: Make the Decision compare t calc vs t critical

 Reject H0 if t > t or if t < -t
 reject H0 the if the p-value  .
 Critical value for
correlation coefficient
 We can calculate the critical value for the correlation coefficient:
 Significance vs. Importance

In large samples, small correlations may be

significant, even if the scatter plot shows little
evidence of linearity. Thus, a significant
(đúng but not qtrong) correlation may lack
practical importance.
 Introduction to
Regression Analysis
 Regression analysis is used to:
 Predict the value of a dependent variable based on the value of independent
variable(s)
 Explain the impact of changes in an independent variable on the dependent
variable
Dependent variable (y): the variable we wish to explain
(response variable)
Independent variable (x): the variable used to explain (predictor variable) the
dependent variable
We want to predict revenue based on marketing budget
Dependent: revenue
Independent: marketing budget
 Simple Linear Regression Model
 The relationship between x and y is described by a
linear function
 Changes in y ( revenue) are assumed to be caused by
changes in x (marketing budget)
 Linear regression population equation model

yi  β 0  β1x i  ε i
 Where 0 and 1 are the population model
coefficients and  is a random error term(other factors
affectign revenue.
 Population Regression Model

Population Random
Population Independent Error
Slope
Y intercept Variable term
Coefficient
Dependent
Variable

y i  β 0  β1x i  ε i
Linear component Random Error
component
 Population Regression Model

y y i  β 0  β1x i  ε i
Observed Value
of y for xi

εi Slope = β1
Predicted Value
Random Error
of y for xi
for this Xi value

Intercept = β0

Xi x
 Sample Regression Model
The sample regression equation provides an
estimate of the population regression line
Estimated Estimate of Estimate of the
(or predicted) the regression regression slope
y value for
observation i intercept
Value of x for

yˆ i  b0  b1x i observation i

The individual random error terms ei have a mean of zero

ei  ( y i - yˆ i )  y i - (b0  b1x i )
 Least Squares Estimators

 b0 and b1 are obtained by finding the values

of b0 and b1 that minimize the sum of the
squared differences between y and ŷ :
min SSE  min  ei2

 min  (y i yˆ i )2

 min  [y i  (b0  b1x i )]2

Differential calculus is used to obtain the

coefficient estimators b0 and b1 that minimize SSE
 Least Squares Estimators
 The slope coefficient estimator is
n

 (x  x)(y  y)
i i
sY
b1  i 1
n
 rxy
sX
 i
(x
i 1
 x) 2

 And the constant or y-intercept is

b0  y  b1x

 The regression line always goes through the mean x, y

 Finding the Least Squares
Equation

 The coefficients b0 and b1 , and other

regression results will be found using a
computer software.
 Hand calculations are tedious
 Statistical routines are built into Excel
 Other statistical analysis software can be used
 Interpretation of the
Slope and the Intercept

 b0 is the estimated average value of y

when the value of x is zero (if x = 0 is in
the range of observed x values)

 b1 is the estimated change in the

average value of y as a result of a one-
unit change in x
 Interpretation of the
Slope and the Intercept
Each extra dependent raises the
mean annual prescription drug
DrugCost = 410 + 550 Dependents cost by $550. An employee with
zero dependents averages $410 in
prescription drugs.

Each extra square foot adds $1.05

to monthly apartment rent. The
Rent = 150 + 1.05 SqFt
intercept is not meaningful because
no apartment can have SqFt = 0.
 Cause and effect
 The relationship between the explanatory variable
and response variable is not causual: one cannot
conclude that the explanatory variable causes a
change in the response variable.
 Consider the following regression equation:
Crime Rate = 0.125 + 0.031 Unemployment Rate
For each unit increase in the unemployment rate, we expect
an increase of .031 in the crime rate, but this does not
mean being out of work causes crime to increase.
 Simple Linear Regression
Example

 A real estate agent wishes to examine the

relationship between the selling price of a home
and its size (measured in square feet)

 A random sample of 10 houses is selected

 Dependent variable (Y) = house price in $1000s
 Independent variable (X) = square feet
Sample Data for House Price
Model
House Price in $1000s Square Feet
(Y) (X)
245 1400
312 1600
279 1700
308 1875
199 1100
219 1550
405 2350
324 2450
319 1425
255 1700
 Graphical Presentation

 House price model: scatter plot

450
400
House Price ($1000s)

350
300
250
200
150
100
50
0
0 500 1000 1500 2000 2500 3000
Square Feet
 Regression Using Excel
 Tools / Data Analysis / Regression
 Excel Output
Regression Statistics
Multiple R 0.76211 The regression equation is:
R Square 0.58082
Adjusted R Square 0.52842 house price  98.24833  0.10977 (square feet)
Standard Error 41.33032
Observations 10

ANOVA
  df SS MS F Significance F
Regression 1 18934.9348 18934.9348 11.0848 0.01039
Residual 8 13665.5652 1708.1957
Total 9 32600.5000      

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580
 Graphical Presentation

 House price model: scatter plot and

regression
450
line
400
House Price ($1000s)

350
Slope
300
250
= 0.10977
200
150
100
50
Intercept 0
= 98.248 0 500 1000 1500 2000 2500 3000
Square Feet

house price  98.24833  0.10977 (square feet)

 Interpretation of the
Intercept, b0

house price  98.24833  0.10977 (square feet)

 b0 is the estimated average value of Y when the

value of X is zero (if X = 0 is in the range of
observed X values)
 Here, no houses had 0 square feet, so b0 = 98.24833
just indicates that, for houses within the range of
sizes observed, $98,248.33 is the portion of the
house price not explained by square feet
 Interpretation of the
Slope Coefficient, b1

house price  98.24833  0.10977 (square feet)

 b1 measures the estimated change in the

average value of Y as a result of a one-
unit change in X
 Here, b1 = .10977 tells us that the average value of a
house increases by 0.10977($1000) = $109.77, on
average, for each additional one square foot of size
 Measures of Variation

 Total variation is made up of two parts:

SST  SSR  SSE
Total Sum of Regression Sum Error Sum of
Squares of Squares Squares

SST   (y i  y)2 SSR   (yˆ i  y)2 SSE   (y i  yˆ i )2

where:
y = Average value of the dependent variable
yi = Observed values of the dependent variable
ŷ = Predicted value of y for the given x value
i i
 Measures of Variation

 SST = total sum of squares

 Measures the variation of the yi values around their
mean, y
 SSR = regression sum of squares
 Explained variation attributable to the linear
relationship between x and y
 SSE = error sum of squares
 Variation attributable to factors other than the linear
relationship between x and y
 Measures of Variation
Y
yi 
 2 y
SSE = (yi - yi )
_
SST = (yi - y)2

y  _2
_ SSR = (yi - y) _
y y

xi X
 Coefficient of Determination, R2
 The coefficient of determination is the portion
of the total variation in the dependent variable
that is explained by variation in the
independent variable
 The coefficient of determination is also called
R-squared and is denoted as R2
SSR SSE
R 2
 1
SST SST

note: 0 R 1
2
 Examples of Approximate
r2 Values
Y
r2 = 1

Perfect linear relationship

between X and Y:
X
r2 = 1
Y 100% of the variation in Y is
explained by variation in X

X
r =1
2
 Examples of Approximate
r2 Values
Y
0 < r2 < 1

Weaker linear relationships

between X and Y:
X
Some but not all of the
Y
variation in Y is explained
by variation in X

X
 Examples of Approximate
r2 Values

r2 = 0
Y
No linear relationship
between X and Y:

The value of Y does not

X depend on X. (None of the
r2 = 0
variation in Y is explained
by variation in X)
 Excel Output
SSR 18934.9348
Regression Statistics
R 2
  0.58082
Multiple R 0.76211 SST 32600.5000
R Square 0.58082
Adjusted R Square 0.52842 58.08% of the variation in
Standard Error 41.33032
house prices is explained by
Observations 10
variation in square feet
ANOVA
  df SS MS F Significance F
Regression 1 18934.9348 18934.9348 11.0848 0.01039
Residual 8 13665.5652 1708.1957
Total 9 32600.5000      

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580
 Estimation of Model
Error Variance
 An estimator for the variance of the population model
error is
n

 i
e 2
SSE
σˆ  s 
2 2
e  i1
n2 n2
 Division by n – 2 is because the simple regression model uses two
estimated parameters, b0 and b1, instead of one

se  s2e is called the standard error of the estimate

 Excel Output
Regression Statistics
Multiple R 0.76211 s e  41.33032
R Square 0.58082
Adjusted R Square 0.52842
Standard Error 41.33032
Observations 10

ANOVA
  df SS MS F Significance F
Regression 1 18934.9348 18934.9348 11.0848 0.01039
Residual 8 13665.5652 1708.1957
Total 9 32600.5000      

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580
 Comparing Standard Errors
se is a measure of the variation of observed y
values from the regression line
Y Y

small se X large se X

The magnitude of se should always be judged relative to the size

of the y values in the sample data

i.e., se = $41.33K is moderately small relative to house prices in

the $200 - $300K range
 Inferences About the
Regression Model

 The variance of the regression slope coefficient

(b1) is estimated by

2 2
s s
s2b1  e
 e

 (xi  x) (n  1)s x
2 2

where:
sb1 = Estimate of the standard error of the least squares slope

SSE
se  = Standard error of the estimate
n2
 Excel Output
Regression Statistics
Multiple R 0.76211
R Square 0.58082
Adjusted R Square 0.52842
Standard Error
Observations
41.33032
10
sb1  0.03297
ANOVA
  df SS MS F Significance F
Regression 1 18934.9348 18934.9348 11.0848 0.01039
Residual 8 13665.5652 1708.1957
Total 9 32600.5000      

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580
 Comparing Standard Errors of
the Slope
Sb1 is a measure of the variation in the slope of regression
lines from different possible samples

Y Y

small Sb1 X large Sb1 X

 Inference about the Slope:
t Test
 t test for a population slope
 Is there a linear relationship between X and Y?
 Null and alternative hypotheses
H0: β1 = 0 (no linear relationship)
H1: β1 0 (linear relationship does exist)
 Test statistic
b1  β1 where:

t b1 = regression slope
sb1 coefficient
β1 = hypothesized slope
sb1 = standard
d.f.  n  2 error of the slope
 Inference about the Slope:
t Test
House Price Estimated Regression Equation:
Square Feet
in $1000s
(x)
(y) house price  98.25  0.1098 (sq.ft.)
245 1400
312 1600
279 1700
308 1875 The slope of this model is 0.1098
199 1100
219 1550
Does square footage of the house
405 2350 affect its sales price?
324 2450
319 1425
255 1700
 Inferences about the Slope:
t Test Example

b1 sb1
H0: β1 = 0 From Excel output:
H1: β1  0   Coefficients Standard Error t Stat P-value
Intercept 98.24833 58.03348 1.69296 0.12892
Square Feet 0.10977 0.03297 3.32938 0.01039

b1  β1 0.10977  0
t  t  3.32938
sb1 0.03297
 Inferences about the Slope:
t Test Example
Test Statistic: t = 3.329
b1 sb1 t
H0: β1 = 0 From Excel output:
H1: β1  0   Coefficients Standard Error t Stat P-value
Intercept 98.24833 58.03348 1.69296 0.12892
d.f. = 10-2 = 8 Square Feet 0.10977 0.03297 3.32938 0.01039

t8,.025 = 2.3060
Decision:
/2=.025 /2=.025 Reject H0
Conclusion:
Reject H0 Do not reject H0 Reject H0
There is sufficient evidence
-tn-2,α/2 0 tn-2,α/2 that square footage affects
-2.3060 2.3060 3.329 house price
 Inferences about the Slope:
t Test Example
P-value = 0.01039
P-value
H0: β1 = 0 From Excel output:
H1: β1  0   Coefficients Standard Error t Stat P-value
Intercept 98.24833 58.03348 1.69296 0.12892
Square Feet 0.10977 0.03297 3.32938 0.01039

This is a two-tail test, so Decision: P-value < α so

the p-value is Reject H0
P(t > 3.329)+P(t < -3.329) Conclusion:
= 0.01039 There is sufficient evidence
(for 8 d.f.) that square footage affects
house price
 Confidence Interval Estimate
for the Slope
Confidence Interval Estimate of the Slope:
b1  t n2,α/2sb1  β1  b1  t n2,α/2sb1
d.f. = n - 2

Excel Printout for House Prices:

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%
Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580

At 95% level of confidence, the confidence interval for

the slope is (0.0337, 0.1858)
 Confidence Interval Estimate
for the Slope
  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%
Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580

Since the units of the house price variable is

$1000s, we are 95% confident that the average
impact on sales price is between $33.70 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
 Confidence intervals and test hyphotheses for
the slope and intercept

NOTE: The test for zero slope is the same as the test for zero
correlation. That is, the t test for zero slope will always yield exactly
the same tcalc as the t test for zero correlation.
 F-Test for Significance
 F Test statistic: MSR
F
where
MSE

SSR
MSR 
k
SSE
MSE 
n  k 1
where F follows an F distribution with k numerator and (n – k - 1)
denominator degrees of freedom

(k = the number of independent variables in the regression model)

ANOVA Table
 Excel Output
Regression Statistics
Multiple R 0.76211
MSR 18934.9348
R Square 0.58082 F   11.0848
Adjusted R Square 0.52842 MSE 1708.1957
Standard Error 41.33032
Observations 10 With 1 and 8 degrees P-value for
of freedom the F-Test
ANOVA
  df SS MS F Significance F
Regression 1 18934.9348 18934.9348 11.0848 0.01039
Residual 8 13665.5652 1708.1957
Total 9 32600.5000      

  Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

Intercept 98.24833 58.03348 1.69296 0.12892 -35.57720 232.07386
Square Feet 0.10977 0.03297 3.32938 0.01039 0.03374 0.18580
 F-Test for Significance

H0: β1 = 0 H0: R2 = 0 Test Statistic:

H1: β1 ≠ 0 H1: R2 > 0 MSR
F  11.08
 = .05 MSE
df1= 1 df2 = 8
Decision:
Critical Reject H0 at  = 0.05
Value:
F = 5.32
 = .05 Conclusion:
There is sufficient evidence that
0 F house size affects selling price
Do not Reject H0
reject H0
F.05 = 5.32
F.05 = FINV(0.05, 1, 8) = 5.32
 Prediction

 The regression equation can be used to

predict a value for y, given a particular x

 For a specified value, xn+1 , the predicted

value is
yˆ n1  b0  b1x n1
 Predictions Using
Regression Analysis
Predict the price for a house
with 2000 square feet:

house price  98.25  0.1098 (sq.ft.)

 98.25  0.1098(2000)

 317.85
The predicted price for a house with 2000
square feet is 317.85($1,000s) = $317,850
 Relevant Data Range
 When using a regression model for prediction,
only predict within the relevant range of data

Relevant data range

450
400
House Price ($1000s)

350
300
250
200
150 Risky to try to
100
extrapolate far
50
0
beyond the range
0 500 1000 1500 2000 2500 3000 of observed X’s
Square Feet
 Confidence Interval for
the Average Y, Given X
Confidence interval estimate for the
expected value of y given a particular xi
Confidence interval for E(Yn1 | Xn1 ) :

 1 (x n1  x)2 
yˆ n1  t n2,α/2se   2
 n  (x i  x) 

Notice that the formula involves the term (x n 1  x)

2

so the size of interval varies according to the distance

xn+1 is from the mean, x
 Prediction Interval for
an Individual Y, Given X
Confidence interval estimate for an actual
observed value of y given a particular xi

Confidence interval for yˆ n1 :

 1 (x n1  x)2 
yˆ n1  t n2,α/2 se 1  2
 n  (x i  x) 

This extra term adds to the interval width to reflect

the added uncertainty for an individual case
 Estimating Mean Values and
Predicting Individual Values
Goal: Form intervals around y to express
uncertainty about the value of y for a given xi
Confidence
Interval for
the expected
Y 
y
value of y,
given xi


y = b0+b1xi

Prediction Interval
for an single
observed y, given xi
xi X
 Estimation of Mean Values:
Example
Confidence Interval Estimate for E(Yn+1|Xn+1)

Find the 95% confidence interval for the mean price

of 2,000 square-foot houses

Predicted Price yi = 317.85 ($1,000s)

1 (x i  x)2
yˆ n1  t n-2,α/2 se   317.85  37.12
n  (xi  x) 2

The confidence interval endpoints are 280.66 and 354.90,

or from $280,660 to $354,900
 Estimation of Individual Values:
Example

Confidence Interval Estimate for yn+1

Find the 95% confidence interval for an individual

house with 2,000 square feet

Predicted Price yi = 317.85 ($1,000s)

1 (Xi  X)2
yˆ n1  t n-1,α/2se 1   317.85  102.28
n  (Xi  X) 2

The confidence interval endpoints are 215.50 and

420.07, or from $215,500 to $420,070
 Finding Confidence and
Prediction Intervals in Excel

Input values


y

Confidence Interval Estimate

for E(Yn+1|Xn+1)

Confidence Interval Estimate


for individual yn+1
 Extrapolation outside the range of X

 Predictions from a regression model are stronger within

the range of sample x values.
 The relationship seen in the scatter plot may not be true
for values far outside observed x range.
 Extrapolation outside the observed range of x is always
tempting but should be approached with caution.
 Summary
 Introduced the linear regression model
 Reviewed correlation and the assumptions of
linear regression
 Discussed estimating the simple linear
regression coefficients
 Described measures of variation
 Described inference about the slope
 Addressed estimation of mean values and
prediction of individual values