Applying a random coefficients logistic model to contractors' decision to bid

Contractors' decision to bid is dependent on many individual characteristics, including some that are unobservable by their competitors. There is natural heterogeneity across contractors in terms of their (i) intrinsic bid/no-bid preferences; and (ii) responses to decision to bid factors. This heterogeneity can be accounted for by applying a random coefficients approach to multiple bid/no-bid responses through logistic modelling. The bid/no-bid data were collected from managers of large and medium-sized contractors in Hong Kong via a designed bidding experiment. Two random coefficients logistic models are developed. Model 1 considers only two groups of decision to bid factors, namely market environment factors (i.e. number of bidders, market conditions) and project-specific factors (i.e. type and size of project). Model 2 extends Model 1 by adding two subject factors (i.e. years of experience, firm size) to study the effect of these individual factors on decision to bid. The results show that there is significant unobserved heterogeneity across contractors and that ignoring its effect results in a downward bias in the parameter estimates of the decision to bid factors. In using this approach contractors can better account for unobserved characteristics of their competitors when formulating their competitive strategies in deciding to bid.