Trend Analysis for Sparse Data

Nurminski, E.A. ORCID: https://orcid.org/0000-0002-7236-6955 & Vorontsov, N.V. (1979). Trend Analysis for Sparse Data. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-79-089

Preview

Text WP-79-089.pdf Download (400kB) | Preview

Abstract

The major theme of this paper is to present some means for an analysis of changes in characteristics of complex systems. Such systems are characterized by a number of parameters which are interdependent. The data available on such systems are sparse in the sense that only a few such systems exist in the real world. Any attempt to choose uniform population with respect to some characteristics will decrease the number of data even more until a statistics approach becomes completely unreasonable.

The alternative approach is based on pattern recognition ideas and uses an idea of separation of different classes of a complex system by multidimensional surfaces. The position of these surfaces demonstrates the trends in the system's development.

The analysis of coal mines with respect to different criteria has been performed as an example.

Actions (login required)

View Item