Académique Documents
Professionnel Documents
Culture Documents
2005 Thomson/South-Western
where:
0 and 1 are called parameters of the model,
2005 Thomson/South-Western
2005 Thomson/South-Western
Slope 1
is positive
xx
2005 Thomson/South-Western
2005 Thomson/South-Western
No Relationship
E(y)
E(y)
Regression line
Intercept
0
Slope 1
is 0
xx
2005 Thomson/South-Western
y b0 b1x
isy
2005 Thomson/South-Western
Estimation Process
Regression Model
y = 0 + 1x +
Regression Equation
E(y) = 0 + 1x
Unknown Parameters
0, 1
b0 and b1
provide estimates of
0 and 1
2005 Thomson/South-Western
Sample Data:
x
y
x1
y1
.
.
.
.
xn
yn
Estimated
Regression Equation
y b0 b1x
Sample Statistics
b0, b1
8
where:
yi = observed value of the dependent variable
for the ith observation
yi =^estimated value of the dependent variable
for the ith observation
2005 Thomson/South-Western
(x x)(y y)
b
(x x)
i
2005 Thomson/South-Western
10
b0 y b1x
where:
xi = value of independent variable for ith
observation
yi = value of dependent variable for ith
_ observation
x = mean value for independent variable
_
y = mean value for dependent variable
n = total number of observations
2005 Thomson/South-Western
11
2005 Thomson/South-Western
12
2005 Thomson/South-Western
13
b1
5
2
4
(xi x)
y 10 5x
2005 Thomson/South-Western
14
Cars Sold
25
20
y = 5x + 10
15
10
5
0
0
2005 Thomson/South-Western
2
TV Ads
15
Coefficient of Determination
SSR
2
2
2
(
y
y
)
(
y
y
)
(
y
y
)
i
i
i i
where:
SST = total sum of squares
SSR = sum of squares due to regression
SSE = sum of squares due to error
2005 Thomson/South-Western
16
Coefficient of Determination
2005 Thomson/South-Western
17
Coefficient of Determination
r2 = SSR/SST = 100/114 = .8772
The regression relationship is very strong; 88%
of the variability in the number of cars sold can be
explained by the linear relationship between the
number of TV ads and the number of cars sold.
2005 Thomson/South-Western
18
rxy (sign of b1 ) r 2
where:
b1 = the slope of the estimated regression
equation
y b0 b1 x
2005 Thomson/South-Western
19
rxy =+ .8772
rxy =
2005 Thomson/South-Western
+.9366
20
2005 Thomson/South-Western
21
and
F Test
Both
Both the
the tt test
test and
and FF test
test require
require an
an estimate
estimate of
of 22,,
the
the variance
variance of
of in
in the
the regression
regression model.
model.
2005 Thomson/South-Western
22
s 2 = MSE = SSE/(n 2)
where:
SSE ( yi y i ) 2 ( yi b0 b1 xi ) 2
2005 Thomson/South-Western
23
An Estimate of
To estimate we take the square root of 2.
The resulting s is called the standard error of
the estimate.
SSE
s MSE
n2
2005 Thomson/South-Western
24
Hypotheses
H 0: 1 0
H a: 1 0
Test Statistic
b1
t
sb1
2005 Thomson/South-Western
25
Rejection Rule
Reject H0 if p-value <
or t < -tor t > t
where:
t is based on a t distribution
with n - 2 degrees of freedom
2005 Thomson/South-Western
26
H a: 1 0
2. Specify the level of significance. = .05
b1
3. Select the test statistic.t
sb1
4. State the rejection rule.
Reject H0 if p-value < .05
or |t| > 3.182 (with
3 degrees of freedom)
2005 Thomson/South-Western
27
b1
5
t
4.63
sb1 1.08
6. Determine whether to reject H0.
t = 4.541 provides an area of .01 in the upper
tail. Hence, the p-value is less than .02. (Also,
t = 4.63 > 3.182.) We can reject H0.
2005 Thomson/South-Western
28
2005 Thomson/South-Western
29
b1 t / 2sb1
is the
margin
of error
b1 is the
point
t / 2 is the t value providing an area
where
estimat
of /2 in the upper tail of a t distribution
or
2005 Thomson/South-Western
30
Rejection Rule
Reject H0 if 0 is not included in
the confidence interval for 1.
1.56 to 8.44
Conclusion
0 is not included in the confidence interval.
Reject H0
2005 Thomson/South-Western
31
Hypotheses
H 0: 1 0
H a: 1 0
Test Statistic
F = MSR/MSE
2005 Thomson/South-Western
32
Rejection Rule
Reject H0 if
p-value <
or F > F
where:
F is based on an F distribution with
1 degree of freedom in the numerator and
n - 2 degrees of freedom in the denominator
2005 Thomson/South-Western
33
H a: 1 0
2005 Thomson/South-Western
34
35
2005 Thomson/South-Western
36
yp t / 2sind
where:
confidence coefficient is 1 - and
t/2 is based on a t distribution
with n - 2 degrees of freedom
2005 Thomson/South-Western
37
Point Estimation
If 3 TV ads are run prior to a sale, we expect
the mean number of cars sold to be:
y =^ 10 + 5(3) = 25 cars
2005 Thomson/South-Western
38
E
CONFIDENCE INTERVAL
xp
3
x bar
2.0
x p -x bar
1.0
(x p -x bar)2
6
7
8
9
10
11
12
13
1.0
2
(x p -x bar)
Variance of y hat
Std. Dev of y hat
t Value
Margin of Error
Point Estimate
Lower Limit
Upper Limit
2005 Thomson/South-Western
4.0
2.1000
1.4491
3.1824
4.6118
25.0
20.39
29.61
39
2005 Thomson/South-Western
40
H
PREDICTION INTERVAL
Variance of y ind
6.76667
Std. Dev. of y ind
2.60128
Margin of Error
8.27845
6
7
2005 Thomson/South-Western
41
2005 Thomson/South-Western
42
Residual Analysis
If the assumptions about the error term appear
questionable, the hypothesis tests about the
significance of the regression relationship and the
interval estimation results may not be valid.
The residuals provide the best information about .
Residual for Observation i
yi yi
Much of the residual analysis is based on an
examination of graphical plots.
2005 Thomson/South-Western
43
2005 Thomson/South-Western
44
Residual
y y
Good Pattern
2005 Thomson/South-Western
45
Residual
y y
Nonconstant Variance
2005 Thomson/South-Western
46
Residual
2005 Thomson/South-Western
47
Residuals
Observation
Residuals
15
-1
25
-1
20
-2
15
25
2005 Thomson/South-Western
48
Residuals
2
1
0
-1
-2
-3
0
2005 Thomson/South-Western
TV Ads
49
End of Chapter 14
2005 Thomson/South-Western
50