Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

 
2007 Annual International CHRIE Conference & Exposition 
132
Table 1 
Pearson Correlation Coefficients 
CAR 
MFEPS 
SFEPS 
OEXFEPS 
CAR 
1 0.1059 0.2314 -0.0865 
(0.4049) 
(0.0658) 
(0.0091) 
MFEPS -0.1059 1 
0.6240 
0.5972 
(0.4049) 
(<.0001) 
(<.0001) 
SFEPS 0.2314 
0.6240 1 
0.6832 
(0.0658) 
(<.0001) 
(<.0001) 
OEXFEPS -0.0865 0.5972 0.6832 
1 
(0.0091) 
(<.0001) 
(<.0001) 
Note
: 
CAR: calculate the dependent variable, CAR, as the sum of residuals from a market model over a 12-month period, 
using a value weighted market index of New York and American Stock Exchange firms. The 12-month 
accumulation period run from nine months prior to the fiscal year-end through three months afterwards. 
MFPSi,t 
= 
(MPSi,t 
– 
MPSi,t-1)/Pi,t-1, 
SFEPSi,t = (SPSi,t – SPSi,t-1)/Pi,t-1, 
OEXFEPSi,t = (OEXPSi,t – OEXPSi,t-1)/Pi,t-1 
Regression analysis results are reported in table 2. Recall that if the estimate of the coefficient of MFEPS is 
not
significantly different from zero, then the implication is that marketing outlays are generating future benefits. 
The test result shows the coefficient is not significantly different from zero at any reasonable level. Moreover, when 
compare the coefficient of MFEPS with that of OEXFEPS, the null hypothesis is rejected at 0.0205 level, indicating 
marketing outlays generate future benefits. To test whether multicollinearity poses serious problems, the values of 
the variance inflation factor (VIF) are calculated. As a rule of thumb, if the resultant values of VIF are less than 5, 
then multicollinearity is not considered as a serious problem. Reported results indicate that all values of VIF are less 
than 5.