scholarly journals IMPROVING NONLINEAR PROCESS MODELING USING MULTIPLE NEURAL NETWORK COMBINATION THROUGH BAYESIAN MODEL AVERAGING (BMA)

2010 ◽  
Vol 9 (1) ◽  
pp. 19-36 ◽  
Author(s):  
Zainal Ahmad ◽  
Tang Pick Ha ◽  
Rabiatul ‘Adawiah Mat Noor
Keyword(s):  
Process Modeling ◽  
Bayesian Model ◽  
Posterior Probability ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Training Data ◽  
Likelihood Method ◽  
Water Tank ◽  
Model Generalization ◽  
Testing Data

Improving model generalization of aggregated multiple neural networks for nonlinear dynamic process modeling using Bayesian Model Averaging (BMA) is proposed in this paper. Using BMA method, the posterior probability of a particular network being the true model is used as the combination weight for aggregating the network despite of using fixed combination weight as the model. The posterior probabilities are calculated using the sum square error (SSE) from the training data on each of the sample time, and tested to the testing data. The selections for the final weight are based on the least SSE calculated when each of the posterior probability is applied to the testing data. The likelihood method is employed for calculating the network error for each input data. Then, it is used to calculate the combination weight for the networks. Two non-linear dynamic system-modeling case studies are selected for this proposed method, which are water tank level prediction and pH neutralization process. Application result demonstrates that the combination using BMA technique can significantly improve model generalization compared to other linear combination approaches.

scholarly journals PEMODELAN AUTOREGRESIF SPASIAL MENGGUNAKAN BAYESIAN MODEL AVERAGING UNTUK DATA PDRB JAWA

2019 ◽  
Vol 3 (3) ◽  
pp. 287-294
Author(s):  
Sarimah Sarimah ◽  
Anik Djuraidah ◽  
Aji H Wigena
Keyword(s):  
Spatial Dependence ◽  
Bayesian Model ◽  
Posterior Probability ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Economic Data ◽  
Explanatory Variables ◽  
Income Per Capita ◽  
Local Revenue ◽  
Per Capita

Economic data always contains spatial effects. Gross Regional Domestic Product (GRDP) in Java is one of economic data that describes spatial dependence between adjacent districts/cities. The method that is suitable for modeling GDRP is spatial regression with spatial dependence on lags that is spatial autoregressive. GDRP prediction used the Bayesian Model Averaging (BMA) method. The ten autoregressive spatial model that have highest posterior probability was chosen to determined the BMA model by posterior probability. The explanatory variables used in this study were (1) mean years of schooling (2) life expectancy (3) income per capita (4) local revenue (5) number of workers (6) district minimum salary. The results showed that the number of workers was chosen as a predictor for the ten models. The model that have highest posterior probability probability is 0.54 which contains five explanatory variables that are mean years of schooling, income per capita, local revenue, number of workers and district minimum salary and the pseudo R2 of the model is 0.696.

scholarly journals Locally Calibrated Probabilistic Temperature Forecasting Using Geostatistical Model Averaging and Local Bayesian Model Averaging

2011 ◽  
Vol 139 (8) ◽  
pp. 2630-2649 ◽  
Author(s):  
William Kleiber ◽  
Adrian E. Raftery ◽  
Jeffrey Baars ◽  
Tilmann Gneiting ◽  
Clifford F. Mass ◽  
...  
Keyword(s):  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Grid Point ◽  
Model Averaging ◽  
Training Data ◽  
Geostatistical Model ◽  
Local Method ◽  
Probabilistic Forecasts ◽  
Temperature Forecasting ◽  
Network Experiment

AbstractThe authors introduce two ways to produce locally calibrated grid-based probabilistic forecasts of temperature. Both start from the Global Bayesian model averaging (Global BMA) statistical postprocessing method, which has constant predictive bias and variance across the domain, and modify it to make it local. The first local method, geostatistical model averaging (GMA), computes the predictive bias and variance at observation stations and interpolates them using a geostatistical model. The second approach, Local BMA, estimates the parameters of BMA at a grid point from stations that are close to the grid point and similar to it in elevation and land use. The results of these two methods applied to the eight-member University of Washington Mesoscale Ensemble (UWME) are given for the 2006 calendar year. GMA was calibrated and sharper than Global BMA, with prediction intervals that were 8% narrower than Global BMA on average. Examples using sparse and dense training networks of stations are shown. The sparse network experiment illustrates the ability of GMA to draw information from the entire training network. The performance of Local BMA was not statistically different from Global BMA in the dense network experiment, and was superior to both GMA and Global BMA in areas with sufficient nearby training data.

Selecting Predictors by Using Bayesian Model Averaging in Bridge Models

2012 ◽  
Author(s):  
Lorenzo Bencivelli ◽  
Massimiliano Giuseppe Marcellino ◽  
Gianluca Moretti
Keyword(s):  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Model Averaging ◽  
Bridge Models

scholarly journals Identification of Mortality Risks in the Advancement of Old Age: Application of Proportional Hazard Models Based on the Stepwise Variable Selection and the Bayesian Model Averaging Approach

Nutrients ◽  
2021 ◽  
Vol 13 (4) ◽  
pp. 1098
Author(s):  
Ewelina Łukaszyk ◽  
Katarzyna Bień-Barkowska ◽  
Barbara Bień
Keyword(s):  
Vitamin D ◽  
Mortality Rate ◽  
Plasma Level ◽  
Variable Selection ◽  
Lymphocyte Count ◽  
Bayesian Model ◽  
Bayesian Model Averaging ◽  
Multivariable Analysis ◽  
Model Averaging ◽  
Nutritional Risk Index

Identifying factors that affect mortality requires a robust statistical approach. This study’s objective is to assess an optimal set of variables that are independently associated with the mortality risk of 433 older comorbid adults that have been discharged from the geriatric ward. We used both the stepwise backward variable selection and the iterative Bayesian model averaging (BMA) approaches to the Cox proportional hazards models. Potential predictors of the mortality rate were based on a broad range of clinical data; functional and laboratory tests, including geriatric nutritional risk index (GNRI); lymphocyte count; vitamin D, and the age-weighted Charlson comorbidity index. The results of the multivariable analysis identified seven explanatory variables that are independently associated with the length of survival. The mortality rate was higher in males than in females; it increased with the comorbidity level and C-reactive proteins plasma level but was negatively affected by a person’s mobility, GNRI and lymphocyte count, as well as the vitamin D plasma level.

Sign in / Sign up

Export Citation Format

Share Document