Regression Models with Categorical Variables 139
fund is defunct (did not survive) and 0 if it survived. The explanatory vari-
ables, their estimated coefficient, and the standard error of the coefficient
using hedge fund data from 1994 to 2003 are given as follows:
Explanatory Variable Coefficient
Standard
Deviation- Return for the first quarter before the end of fund
performance. –1.47 0.36 - Return for the second quarter before the end of fund
performance. –4.93 0.32 - Return for the third quarter before the end of fund
performance. –2.74 0.33 - Return for the fourth quarter before the end of fund
performance. –3.71 0.35 - Standard deviation for the year prior to the end of
fund performance. 17.76 0.92 - Number of times in the final three months the fund’s
monthly return fell below the monthly median of all
funds in the same primary category. 0.00 0.33 - Assets of the fund (in billions of dollars) estimated at
the end of performance. –1.30 –7.76
Constant term –0.37 0.07
For only one explanatory variable, the sixth one, the coefficient is not
statistically significant from zero. That explanatory variable is a proxy for
peer comparison of the hedge fund versus similar hedge funds. The results
suggest that there is a lower probability of the demise of a hedge fund if
there is good recent performance (the negative coefficient of the first four
variables above) and the more assets under management (the negative coef-
ficient for the last variable above). The greater the hedge fund performance
return variability, the higher the probability of demise (the positive coeffi-
cient for the fifth variable above).
logit regression Model
As with the probit regression model, the logit regression model is a non-
linear regression model where the dependent variable is a binary variable
and the predicted values are between 0 and 1. The predicted value is also a
cumulative probability distribution. However, rather than being a standard