About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
AAAI 1990
Conference paper
Obtaining Quantitative Predictions From Monotone Relationships
Abstract
Quantitative predictions are typically obtained by characterizing a system in terms of algebraic relationships and then using these relationships to compute quantitative predictions from numerical data. For real-life systems, such as mainframe operating systems, an algebraic characterization is often difficult, if not intractable. This paper proposes a statistical approach to obtaining quantitative predictions from monotone relationships - non-parametric interpolative-prediction for monotone functions (NIMF). NIMF uses monotone relationships to search historical data for bounds that provide a desired level of statistical confidence. We evaluate NIMF by comparing its predictions to those of linear least-squares regression (a widely-used statistical technique that requires specifying algebraic relationships) for memory contention in an IBM computer system. Our results suggest that using an accurate monotone relationship can produce better quantitative predictions than using an approximate algebraic relationship.