Appendix A Appendix

(1)

Appendix

Appendix A

TABLE 9 A

Test for Formality for Conglomerates

Kolmogorov-Smirnov(a) Shapiro-Wilk

_Statistic _df _Sig. _Statistic _df _Sig. Return on Assets _.099 ₆₇ _.170 _.971 ₆₇ _.126 Virtual Return on Assets _.099 ₆₆ _.181 _.965 ₆₆ _.059 Return on Cash Assets .104 67 .070 .929 67 .001 Virtual Return on Cash

Assets .103 66 .077 .937 66 .002 Return on Sales .101 67 .084 .901 67 .000 Virtual Return on Sales _.127 ₆₆ _.010 _.948 ₆₆ _.008 Herfindahl index .101 68 .080 .960 68 .029 Portfolio strategy _.524 ₆₈ _.000 _.375 ₆₈ _.000 Leverage .079 67 .200(*) .972 67 .129 Virtual leverage _.105 ₆₆ _.066 _.922 ₆₆ _.000 Geographic diversification .525 61 .000 .371 61 .000 Virtual Geographic Diversifi _.386 ₆₆ _.000 _.624 ₆₆ _.000 Herfindahl Index 2 .170 68 .000 .884 68 .000

* This is a lower bound of the true significance. a Lilliefors Significance Correction

TABLE 9 B

Test for Normality for One-Segment Firms

Kolmogorov-Smirnov(a) Shapiro-Wilk

_Statistic _df _Sig. _Statistic _df _Sig. Return on Assets .138 22 .200(*) .970 22 .707 Virtual Return on Assets _.133 ₂₅ _.200(*) _.967 ₂₅ _.576 Return on Cash Assets .155 22 .185 .952 22 .347 Virtual Return on Cash

Assets .118 25 .200(*) .963 25 .480 Return on Sales .105 22 .200(*) .964 22 .577 Virtual Return on Sales _.154 ₂₅ _.130 _.932 ₂₅ _.097 Portfolio strategy .521 25 .000 .384 25 .000 Leverage _.134 ₂₄ _.200(*) _.952 ₂₄ _.302 Virtual leverage .104 23 .200(*) .981 23 .927 Geographic diversification _.469 ₂₁ _.000 _.533 ₂₁ _.000 Virtual Geographic diversifi .521 25 .000 .384 25 .000

* This is a lower bound of the true significance. a Lilliefors Significance Correction

(2)

Appendix B

Most of the assumptions pertain to error. For instance, it is assumed that the residuals

are normally distributed, they have a mean of zero and a constant variance across levels of

independent variables, they are independent across all subjects, and they are not

systematically related (correlated) to the independent variables.

A histogram and a normal probability plot displays the extent of to which the residuals

follow a normal distribution. A histogram of the standardized residuals with a superimposed

normal curve displays that the residuals do not adhere perfectly to a normal distribution;

however they do not appear to depart significantly.

FIGURE 3 A

(3)

FIGURE 3 B

(4)

FIGURE 3C

Histogram of Standardized Residuals for the dependent variable ROS

(5)

FIGURE 4 A

Normal P-P plot of standardized residual for the dependent variable ROA

FIGURE 4 B

(6)

FIGURE 4 C

Normal P-P plot of standardized residual for the dependent variable ROS

(7)

FIGURE 5 A

Scatter plot for the dependent variable ROA

FIGURE 5 B

(8)