Predicting with sparse data

It is well known that effective prediction of project cost related factors is an important aspect of software engineering. Unfortunately despite exten sive research over more than 30 years, this remains a significant problem f or many practitioners. A major obstacle is the absence of reliable and syst ematic historic data, yet this is a sine qua non far almost all proposed me thods: statistical, machine learning or calibration of existing models. In this paper we describe our sparse data method (SDM) based upon a pairwise c omparison technique and Saaty's Analytic Hierarchy Process. Our minimum dat a requirement is a single known point. The technique is supported by a soft ware tool known as DataSalvage. We show, for data from two companies, how o ur approach - based upon expert judgement - adds value to expert judgement by producing significantly more accurate and less biased results. A sensiti vity analysis shows that our approach is robust to pairwise comparison erro rs. We then describe the results of a small usability trial with a practisi ng project manager. From this empirical work we conclude that the technique is promising and may help overcome some of the present barriers to effecti ve project prediction.