Forecasting based on Bayesian type models
Journal Title: INTERNATIONAL JOURNAL OF COMPUTERS & TECHNOLOGY - Year 2016, Vol 15, Issue 3
Abstract
A review of some Bayesian data analysis models is proposed, namely the models with one and several parameters. A methodology is developed for probabilistic models construction in the form of Bayesian networks using statistical data and expert estimates. The methodology provides a possibility for constructing high adequacy probabilistic models for solving the problems of classification and forecasting. An integrated dynamic network model is proposed that is based on combination of probabilistic and regression approaches; the model is distinguished with a possibility for multistep forecasts estimation. The forecast estimates computed with the dynamic model are compared with the results achieved with logistic regression combined with multiple regression. The best results were achieved in this case with the combined dynamic net model.Â
Authors and Affiliations
Peter Bidyuk, Aleksander Peter Gozhjy, Alexandr T rofymchuk
Keywords
Related Articles
- EP ID EP650782
- DOI 10.24297/ijct.v15i3.1672
- Views 117
- Downloads 0
Challenges and Security Issues in Future IT Infrastructure Components
Over the past 2 decades, the information technology infrastructure has gone through an exponential change with the introduction and evolution of new technologies and trends. Organizations previously having their data on-...
Performance Comparison of IMABN-2 and MALN-2 in Faulty and Non-Faulty Network Conditions
In this paper, we have compared the performance of IMABN-2 and MALN-2. To compare the performance, data packets were passed via both networks in faulty and non-faulty environments. Results show that MALN-2 performs bette...
Optimal Electric-Power Distribution and Load-Sharing on Smart-Grids: Analysis by Artificial Neural Network
This study refers to developing an electric-power distribution system with optimal/suboptimal load-sharing in the complex and expanding metro power-grid infrastructure. That is, the relevant exercise is to in...
A Fully Automated Method for Noisy cDNA Microarray Image Quantification
DNA microarray is an innovative tool for gene studies in biomedical research, and its applications can vary from cancer diagnosis to human identification. Image processing is an important aspect of microarray experiments...
STUDY OF INCREASE IN GLOBAL COMPONENTS IN HYBRID WAVELETS ON DATA COMPRESSION
This paper presents a hybrid wavelet transform technique which studies the effect of global components on the quality of image compression. Hybrid wavelet transform is generated using two different component orthogonal t...