In the first two essays of this dissertation, I construct a dynamic stochastic production frontier incorporating the sluggish adjustment of inputs, measure the speed of adjustment of output in the short-run, and compare the technical efficiency estimates from such a dynamic model to those from a conventional static model that is based on the assumption that inputs are instantaneously adjustable in a production system. I provide estimation methods for technical efficiency of production units and the speed of adjustment of output for cases when they are time-invariant and when they vary with time. I also apply the methods to a panel dataset on private manufacturing establishments in Egypt.

The dynamic frontiers with time-invariant and time-varying technical efficiency are estimated using the System Generalized Method of Moments estimator and the Generalized Least Squares estimator with instrumental variables, respectively. The results for the Egyptian private manufacturing sectors show that the speed of adjustment of output is significantly lower than unity, the static model underestimates technical efficiency on average, and the dynamic model captures more variation in the time pattern of technical efficiency. Further, the ranking of production units based on their technical efficiency measures changes when the lagged adjustment process of inputs is taken into account.

In another essay, I characterize a class of rules for decision-making under the type of non-probabilistic uncertainty that was first axiomatically analyzed by Arrow and Hurwicz (1972). In this framework, the agent knows the possible states of the world and the outcome of each of her actions for each state, but does not have any information about the probabilities with which each state occurs. The decision-making rules characterized in this essay focus on the outcome(s) which occupy the middle position(s), when all outcomes of an action under different states of the world are arranged according to the agent's preference ordering defined over the outcomes. The existing literature in the Arrow-Hurwicz framework has mainly considered `max'-based or `min'-based rules and their variants, which reflect rather extreme forms of optimism or pessimism on the part of an agent. In contrast, the results of this essay characterize a decision rule that reflects a more `balanced' attitude of the agent.