Building a software cost estimation model based an categorical data

This paper explores the possibility of generating a multi-organisational so ftware cost estimation model by analysing the software cost data collected by the International Software Benchmarking Standards Group. This database c ontains data about recently developed projects characterised mostly by attr ibutes of categorical nature such as the project business area, organisatio n type, application domain and usage of certain tools or methods. The gener ation of the model is based on a statistical technique which has been propo sed as alternative to the standard regression approach, namely the categori cal regression or regression with optimal scaling. This technique starts wi th the quantification of the qualitative attributes (expressed either on no minal or ordinal scale), that appear frequently within such data, and proce eds by using the obtained scores as independent variables of a regression m odel The generated model is validated by measuring certain indicators of ac curacy.