When may the accuracy of expert estimation be improved by using historical data?

Abstract

As expert estimation is the estimation strategy most frequently applied to software projects today, it is important to focus the research on effort estimation methods on it. This is the estimation method used in Agile contexts so we are interested in deeply understanding the value of historical data in this context, in particular when the project domains and the technological environments are new to the team, and when the teams -with little experience in Agile contexts- have recently been created.