Professor Basil Kalymon, in his recent article [1], correctly asserts that the appropriate variance for the portfolio selection model, reflecting portfolio risk, should measure the risks created “not only by the inherent fluctuations of returns, but also by the decision-maker's lack of perfect information about the parameters of his model” [1, p. 560]. He identifies a significant failing of portfolio selection models, that “the question of estimating the required parameters in the models has been largely sidestepped … by assuming that the parameter values are known” [1, p. 559]. By showing that error in estimating security returns is an important component of risk in portfolio selection models, Professor Kalymon has taken portfolio theory an important step forward.