Application of Logistic Regression to Supporting Contractors Decision

Abstract

The bid/no bid decision is one of the first and most important decisions to make by the contractor when seeking a construction contract. To facilitate the contractor’s reasoning by limiting randomness that may lead to mistakes decision support models are frequently applied. This paper presents possible applications of a logistic regression to support bidding decisions. On the basis of a set of data, the model was to suggest either participation in the bid or resignation from it. Methods of estimation and measurement of goodness of fit are presented in the paper. The results, 79,55% of correctly classified cases, confirmed the usability of logistic regression in solving the problem.