Optimization of Forecasting Models for Testing Blood Samples by Estimation of Tiol-Disulfid Diagrams. Nina Kondrashova, Andriy Pavlov, Yaroslav Pavlov. IWIM, Prague, 2007.

 Article (in pdf)

Abstract. Constructing the forecast models for values of tiol-disulfide ratios in blood samples in several measurement points is considered for decreasing the time of patients’ examination. Models are obtained by the GMDH algorithms. The number of examination measurement points in which models have feasible error is maximized by optimization of initial data sample division, by adaptive forecast, and sequential use of selection criteria. The criteria and results of numeric experiments are given. Application of difference models with an adaptive forecast and algorithm of modeling with the twostage division of initial sample are effective for the variables forecast.

Keywords. Forecasting model, GMDH algorithm, sample division, adaptive forecast, tiol-disulfide ratio

References.