An inverse hyperbolic sine transformation is:
IHS = log(z + sqrt(z^z + 1))
It is an alternative to log transformations when some of
the variables take on zero or negative values and as
an alternative to the Box-Cox when variables are zero or
negative.
See, for example:
Layton, David F, 2001. "Alternative Approaches for Modeling
Concave Willingness to Pay Functions in Conjoint Valuation,"
American Journalof Agricultural Economics, vol. 83(5), 1314-20.
Burbidge, John B., Lonnie Magee and A. Leslie Robb. 1988 "Alternative
Transformations to Handle Extreme Values of the Dependent Variable."
Journal of the American Statistical Association, vol. 83, 123-127.
Scott
----- Original Message -----
From: [email protected]
Date: Wednesday, March 30, 2005 1:40 pm
Subject: st: how to use inverse hyperbolic sine function
Dear users,
I need to use inverse hyperbolic sine function to transform my
dependentvariables in a regression, which, honestly, I don't
understand very much.
I failed to find the relevant commands to perform this technique
in Stata
7.0. Does anybody have a relevant program which I can use as a
reference,and have suggestions about interpreting the coefficient
for varaibles like
gender? Thanks a lot.
Ying
