Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Since Fishburn and Porter (1976), it has been known that a first-order dominant shift in the distribution of random returns of an asset does not necessarily induce a risk-averse decision maker to increase his holdings of that improved asset. To obtain the desired comparative statics result, one has to further restrict the class of changes in the distribution. In this article, we propose the "monotone probability ratio" criterion which is more general than the "monotone likelihood ratio" criterion currently used in the literature. Copyright 1995 by Kluwer Academic Publishers