Chemistry
5 years ago

New weight-based model selection criterion improves accuracy in time series analysis.

International Journal of Applied Mathematics & Statistics/International journal of applied mathematics and statistics
Information Criterion for Selecting Competing Models Using Different Weighting Schemes
Wilhemina Adoma Pels, Sampson Twumasi-Ankrah, Atinuke O. Adebanji

Paper Summary

A new method for choosing the best model in time series analysis was developed by combining different information criteria. By using a weighted approach, the new method outperformed traditional criteria like AIC and SIC. The results showed that the new method, especially when combined with the Hannan Quinn criterion, was the most effective in selecting the most accurate model.

Related papers

Theme background image
Computer Science
8 years ago

New study reveals best method for predicting future trends accurately!

Neural Networks
Multicollinearity
Regularization Methods
1 view
Theme background image
Mathematics
8 years ago

New information criteria revolutionize model selection in linear regression.

Regularization Methods
Bayesian Networks
Model-Based Clustering
1 view
Theme background image
Engineering
5 years ago

New autoregression criterion adapts to data, outperforming traditional model selection methods.

System Identification
Blind Source Separation
Neural Networks
1 view