Değerli Araştırmacılar,
Tennessee Üniversitesi Business Analytics and Statistics Bölümü öğretim üyesi Prof.Dr. Hamparsum Bozdoğan, 23 Mayıs 2019 tarihinde İstanbul Üniversitesi/İşletme Fakültesi/Mavi Salonda saat 14.30’da “NOVEL DIMENSION REDUCTION TECHNIQUES FOR HIGH-DIMENSIONAL DATA USING INFORMATION COMPLEXITY” adlı çalışmasını sunacaktır.
Sizleri de aramızda görmekten mutluluk duyarız.
İ.Ü. Işletme Fakültesi

NOVEL DIMENSION REDUCTION TECHNIQUES FOR HIGH-DIMENSIONAL DATA USING INFORMATION COMPLEXITY*

Prof. Dr. HAMPARSUM BOZDOGAN, PhD
McKenzie Professor
Department of Business Analytics and Statistics
The University of Tennessee
Knoxville, TN 37996 U.S.A.
ABSTRACT

This presentation introduces and develops two computationally feasible intelligent feature extraction techniques that address potentially daunting statistical and combinatorial problems. The first part of the talk employs a three-way hybrid of Probabilistic Principal Component Analysis (PPCA) to reduce the dimensionality of the dependent variables, Multivariate Regression (MVR) models that account for misspecification of the distributional assumption to determine a predictive operating model for glass composition for automobiles, and the Genetic Algorithm (GA) as the optimizer, along with the misspecification-resistant form of Bozdogan's Information Measure of Complexity (ICOMP) as the fitness function. The second part of the talk is devoted to dimension reduction via a novel Adaptive Elastic Net (AEN) regression model. AEN model is used to reduce the dimension of a Japanese stock index called TOPIX as a response to build a best predictive model when we have a “large p, small n" problem. Our results show the remarkable dimension reduction in both of these real-life examples of wide data sets, which demonstrates the versatility and the utility of the two proposed novel statistical data modeling techniques.

KEYWORDS: Dimension Reduction; Information Complexity; Probabilistic Principal Component Analysis (PPCA), Genetic Algorithm; Multivariate Regression Model; Adaptive Elastic Net
Model; Predictive Modeling.

*Joint work with Esra Pamukcu.