ARRAY(0x1009d2148)

This thesis addresses three major issues in data mining regarding feature subset selection in large dimensionality domains, plausible reconstruction of incomplete data in cross-sectional applications, and forecasting univariate time series. For the automated selection of an optimal subset of...

We present an update of mim, a program for managing multiply im- puted datasets and performing inference (estimating parameters) using Rubin’s rules for combining estimates from imputed datasets. The new features of particular importance are an option for estimating the Monte Carlo error (due...

This paper focuses on the construction of a composite indicator for the knowledge based economy using imprecise data. Specifically, for some indicators we only have information on the bounds of the interval within which the true value is believed to lie. The proposed approach is based on a...