We consider the problem faced by a seller in determining an optimal price to quote
in response to a Request for Quote (RFQ) from a prospective buyer. The optimal price
is determined by maximizing expected profit, given the underlying seller costs of the bid items and a computed probability of winning the bid as a function of price and other bid features such as buyer characteristics and the degree of competition. An entropy-based information-gain metric is used to quantify the contribution of the extracted features to predicting the win/loss label. A naive Bayes classification model is developed to predict the bid outcome (win or loss) as a function of these features. This model naturally generates the win probability as a function of bid price required to compute the optimal price. Results obtained by applying this model to a database of bid transactions involving computer sales demonstrate statistically significant lift curves for predicting bid outcome. A method for creating additional synthetic bids to improve computation of the win probability function is demonstrated. Finally, the computed optimal prices generated via this approach are compared to the actual bid prices approved by human pricing experts.
By: Richard D. Lawrence
Published in: RC22473 in 2002
LIMITED DISTRIBUTION NOTICE:
This Research Report is available. This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. It has been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g., payment of royalties). I have read and understand this notice and am a member of the scientific community outside or inside of IBM seeking a single copy only.
Questions about this service can be mailed to reports@us.ibm.com .