Consumer Surveys
data from survey questions
Observational Research
data from observed behavior
Consumer Clinics
data from laboratory experiments
Market Experiments
data from real market tests
Regression Analysis
Year 1 2 3 4 5 6 7 8 9 10 X 10 9 11 12 11 12 13 13 14 15 Y 44 40 42 46 48 52 54 58 56 60
Scatter Diagram
Regression Analysis
Regression Line: Line of Best Fit
Regression Line: Minimizes the sum of the squared vertical deviations (et) of each point from the regression line. Ordinary Least Squares (OLS) Method
Yt a bX t et Yt Yt
(X
t 1 n t 1
X )(Yt Y )
t
(X
a Y bX
X)
Xt
10 9 11 12 11 12 13 13 14 15 120
Yt
44 40 42 46 48 52 54 58 56 60 500
Xt X
-2 -3 -1 0 -1 0 1 1 2 3
n
Yt Y
-6 -10 -8 -4 -2 2 4 8 6 10
( X t X )(Yt Y )
12 30 8 0 2 0 4 8 12 30 106
( X t X )2
4 9 1 0 1 0 1 1 4 9 30
n 10
X
t 1 n
X t 120
t 1
Yt 500
t 1
(X
t 1 n
X ) 2 30 X )(Yt Y ) 106
106 3.533 b 30
a 50 (3.533)(12) 7.60
X t 120 12 n 10
Yt 500 50 10 t 1 n
(X
t 1
X t 120 12 n 10
X
t 1
120
2
Y
t 1
500
Yt 500 50 10 t 1 n
(X
t 1 n
X ) 30 X )(Yt Y ) 106
106 3.533 b 30
(X
t 1
a 50 (3.533)(12) 7.60
Tests of Significance
Standard Error of the Slope Estimate
(Yt Y ) 2 et2
sb
(n k ) ( X t X )
(n k ) ( X t X ) 2
Tests of Significance
Example Calculation
Time
1 2 3 4 5 6 7 8 9 10
Xt
10 9 11 12 11 12 13 13 14 15
Yt
44 40 42 46 48 52 54 58 56 60
Yt
42.90 39.37 46.43 49.96 46.43 49.96 53.49 53.49 57.02 60.55
et Yt Yt
1.10 0.63 -4.43 -3.96 1.57 2.04 0.51 4.51 -1.02 -0.55
et2 (Yt Yt )2
1.2100 0.3969 19.6249 15.6816 2.4649 4.1616 0.2601 20.3401 1.0404 0.3025 65.4830
( X t X )2
4 9 1 0 1 0 1 1 4 9 30
e (Yt Yt )2 65.4830
t 1 2 t t 1
(X
t 1
X ) 30
2
sb
(Y Y ) ( n k ) ( X X )
2 t t
Tests of Significance
Example Calculation
et2 (Yt Yt ) 2 65.4830
t 1 t 1 n n
( X t X ) 2 30
t 1
)2 (Yt Y
Tests of Significance
Calculation of the t Statistic
b 3.53 t 6.79 sb 0.52
Tests of Significance
Decomposition of Sum of Squares
Total Variation = Explained Variation + Unexplained Variation
Y )2 (Y Y )2 (Yt Y ) (Y t t
2
Tests of Significance
Coefficient of Determination
R2 Explained Variation TotalVariation (Yt Y )2 (Y Y )2
Tests of Significance
Coefficient of Correlation
r R2 withthe signof b
1 r 1
r 0.85 0.92
Model:
Y a b1 X 1 b2 X 2
bk ' X k '
R 2 1 (1 R 2 )
Durbin-Watson Statistic
Test for Autocorrelation
(et et 1 ) 2
t 2 n
et2
t 1
If d = 2, autocorrelation is absent.
Power Function:
b QX a ( PX1 )( PYb2 )
Estimation Format:
ln QX ln a b1 ln PX b2 ln P Y
Regression Output
Regression Output