3,673 works, 150 years of economic thought. Each one summarized and searchable, with cited passages inside.
An operating policy can remain optimal despite errors in its coefficients while its returns—and the ranking of alternative policies—change. J. K. Sengupta, C. Millham, and Gerhard Tintner make this distinction central to their study of stochastic linear programming. Their numerical example preserves the winning selection under ten-percent coefficient errors yet allows inferior selections to exchange places, giving concrete force to the question of what, exactly, is stable. They then compare the variability of best, second-best, and third-best returns, exploring a conditional trade-off between higher returns and lower dispersion through an Iowa farm example. The article offers readers a precise way to distinguish persistence of an optimal policy from stability of its payoff, while its variance argument requires closer scrutiny of the assumptions supporting that trade-off.