Professional Documents
Culture Documents
Multiple Regression
Multiple Regression
Multiple Regression
JOHN
LOUCKS
& Updated by
SPIROS
VELIANITIS
where:
0, 1, 2, . . . , p are the parameters, and
is a random variable called the error term
^
y = b0 + b1x1 + b2x2 + . . . + bpxp
Estimated Multiple
b0, b1, b2, . . . , bp Regression Equation
provide estimates of yˆ b0 b1 x1 b2 x2 ... bp x p
0, 1, 2, . . . , p Sample statistics are
b0, b1, b2, . . . , bp
min ( yi yˆ i )2
where
y = annual salary ($1000)
x1 = years of experience
x2 = score on programmer aptitude test
Least Squares
Input Data Output
x1 x2 y Computer b0 =
Package b1 =
4 78 24 for Solving
7 100 43 b2 =
Multiple
. . .
Regression R2 =
. . .
3 89 30 Problems etc.
b1 = 1.404
b2 = 0.251
i
( y y ) 2
= i
( ˆ
y y ) 2
+ i i
( y ˆ
y ) 2
where:
SST = total sum of squares
SSR = sum of squares due to regression
SSE = sum of squares due to error
R2 = SSR/SST
R2 = 500.3285/599.7855 = .83418
n1
Ra2 2
1 (1 R )
np1
2 20 1
R 1 (1 .834179)
a .814671
20 2 1
Hypotheses H 0: 1 = 2 = . . . = p = 0
Ha: One or more of the parameters
is not equal to zero.
Hypotheses H0 : i 0
H a : i 0
bi
Test Statistics t
sbi
where:
y^ = annual salary ($1000)
x1 = years of experience
x2 = score on programmer aptitude test
x3 = 0 if individual does not have a graduate degree
1 if individual does have a graduate degree
x3 is a dummy variable
Not significant
Highest
Degree x1 x2
Bachelor’s 0 0
Master’s 1 0
Ph.D. 0 1
2
Residuals
Standard
0
0 10 20 30 40 50
-1
-2
Predicted Salary
0 1 x1 2 x2 p x p
e
E( y ) 0 1 x1 2 x2 p x p
1 e
Interpretation of E(y) as a
Probability in Logistic Regression
If the two values of y are coded as 0 or 1, the value
of E(y) provides the probability that y = 1 given a
particular set of values for x1, x2, . . . , xp.
E( y ) estimate of P( y 1|x1 , x2 , , x p )
b0 b1 x1 b2 x2 bp x p
e
yˆ b0 b1x1 b2 x2 bp x p
1 e
Odds 95% CI
Predictor Coef SE Coef Z p Ratio Lower Upper
Log-Likelihood = -60.487
Test that all slopes are zero: G = 13.628, DF = 2, P-Value = 0.001
Hypotheses H0: 1 = 2 = 0
Ha: One or both of the parameters
is not equal to zero.
Test Statistics z = bi/sb
i
Odds Ratio
odds 1
Odds Ratio
odds 0
Estimated Probabilities
Annual Spending
$1000 $2000 $3000 $4000 $5000 $6000 $7000
Computed
earlier
Comparing Odds
Suppose we want to compare the odds of making a
$200 purchase for customers who spend $2000 annually
and have a Simmons credit card to the odds of making a
$200 purchase for customers who spend $2000 annually
and do not have a Simmons credit card.
.4099
estimate of odds 1 .6946
1 - .4099
.1880
estimate of odds 0 .2315
1 - .1880
.6946
Estimate of odds ratio 3.00
.2315