Despite recent advances incorporating uncertainty into the theory of the firm, major problems remain which mar the theory and prevent its practical application. This paper explores and offers solutions to two such problems: the form of the neoclassical firm's objective function under uncertainty, and the derivation of equations for investment and factor inputs from the maximization of a properly chosen objective function. The expression adopted in Section I as the best available objective function for the neoclassical firm is derived from the well-known Lintner-Mossin-Sharpe (LMS) market valuation equation (see the articles by John Lintner (1965a), Jan Mossin, and William Sharpe). However, the major point of this section is that the LMS equation, as presently written, unnecessarily limits the scope of the analysis of optimal firm decisions, particularly investment decisions. Implicit in the currently accepted version of the equation is the assumption that todlay's decisions affect firm cash flow only in the present period. Truly dlynamic problems such as those attacked in the investment analysis pioneered by Robert EIisner and Robert Strotz cannot, therefore. he handlledT'he situation can be remedied, but we will see that this is more difficult in some cases than others. It will depend on one's choice of assumptions about the randomness of future values of the market parameters: the riskless rate of interest and the market price of risk. In the second part of this paper, a revised version of the LMS valuation equation is used to derive optimal investment, labor input, and production policies for models that have been extensively studied under certainty, but which have been treated inadequately or not at all under uncertainty. In particular, we will cover 1) the standard neoclassical case where the firm faces constant factor costs and constant returns to scale, and 2) the EisnerStrotz model, where the cost of investment goods is no longer constant, but a function of the rate of investment.