From Product Description to Cost: A Practical Approach [electronic resource] : Volume 2: Building a Specific Model / by Pierre Foussier.

Population and Sample -- From the Sample to the Population -- Describing the Population -- Typical Distributions -- Data Analysis Precedes the Search for a Specific Model -- Data Analysis on One Variable Only -- Data Analysis on Two Variables -- Simultaneous Data Analysis on J + 1 Quantitative Variables -- Working with Qualitative Variables -- Finding the Dynamic Center of a Multi-Variables Sample -- Finding the Center of the Cost Distribution for Choosing a Metric -- Looking for the Dynamic Center: The Bilinear Cases -- Using Several Quantitative Parameters: The Linear Cases -- Using Qualitative Variables -- Non-Linear Relationships -- Studying the Residuals Is as Important as Finding the Formula -- Studying the Additive Residuals -- The Other Residuals -- Building a Specific Model -- From Sample to Population -- Building the Model.

Parametric cost estimating, when properly used, is a very powerful and cost effective technique. Unfortunately, this technique is often misunderstood and rejected by many potential users for whom it could be a beneficial tool. The development of an agreement on how to use it and how to recognise its limitations is a major benefit to the cost estimating community. From Product Description to Cost: A Practical Approach: • presents different ways of parametrically forecasting costs, and the advantages and disadvantages of these methods by using real examples from the mechanical, software and building industries; • discusses most of the mathematical procedures which are useful for parametrically forecasting costs; and, • introduces the judgement needed to audit the ways in which these forecasting techniques are used, firstly as a process, and secondly as a tool to generate estimates. Volume 2: Building a Specific Model is divided into nine parts and focuses on the building of ‘specific’ cost estimating models. These models explicitly refer to a ‘product family’, and, as opposed to ‘general’ models, any company (if properly organised) can build these models. Understanding the procedures is key in creating successful specific models; for this reason all these procedures are fully described in this volume. Classical methods and new ones (such as the "bootstrap") are described and illustrated. The Decision Engineering series focuses on the foundations and applications of tools and techniques related to decision engineering, and identifies their relevance in ‘engineering’ decisions. The series provides an aid to practising professionals and applied researchers in the development of tools for informed operational and business decision making, within industry, by utilising distributed organisational knowledge.