Sentence examples for model selection used from inspiring English sources

The Bayesian information criterion (BIC) is another criterion for model selection, used to correct the overfitting nature of the AIC [10].

Final model selection used a "leave-one-out, cross-validation" procedure to reduce over-fitting, a minimum data-to-predictor ratio (10 1), and log-likelihood ratio improvement criterion to ensure parsimony.

Model selection used a stepwise backward-forward procedure.

Main effects and interactions were tested and model selection used Akaike's information criterion (AIC).

The model selection used the backward stepwise method (likelihood ratios), and variables at a P value less than 0.05 were retained in the model as independent variables.

Model selection used log-likelihood from 10-fold cross-validation, the Bayesian information criterion (BIC) of the likelihood of the full data, and BIC of 100 bootstrap replicates of the full data to assess stability.

In theory, at infinite sample size, model selection using BIC selects the true model if it is included in the search space [56].

Model selection using AIC criteria allowed us to identify the time period over which to consider management practices that best explain plant patterns.

A stepwise model selection analysis was performed, yielding the same markers that were selected by stepwise model selection using an ordinary Cox model (data not shown).

Performing model selection using Bayes factors is particularly useful when dealing with a few models only.

We therefore chose to perform model selection using the AIC.