Using Multi-Logistic Distribution and Stochastic Goal Programming for Estimating the Human Resources for Development Strategy

The main goal of this paper is to anticipate the expected estimates of the human resources to build up a development strategy of employment. Three methodological approaches were used to achieve this goal, the logistic distribution, Bayesian technique and the stochastic goal programming (SGP). The multi-logistic regression (MLR) model was used to analyze the relationship between a multi-categorical outcome of human capital and their human resources. Using the SGP approach, the MLR solution gave percentages of correct allocations for the human resources among the human variables as well as it provided the optimal estimates for the parameters of the model. An alternative and richer MLR with matched case- control outcomes model was built for each of the human force categories applying the Bayesian approach to construct the matched model and applying the SGP solution gave more information about the parameters’ estimates of the MLR model. Finally, a SGP model was constructed, inversely, to project the human resources categories satisfying the development goals of labor productivity for the planned output pattern.