We investigate the Bayesian Information Criterion (BIC) for variable selection in models for censored survival data. Kass and Wasserman (1995, Journal of the American Statistical Association 90, 928-934) showed that BIC provides a close approximation to the Bayes factor when a unit-information prior on the parameter space is used. We propose a revision of the penalty term in BIC so that it is d...

fuzzy decision tree (fdt) classifiers combine decision trees with approximate reasoning offered by fuzzy representation to deal with language and measurement uncertainties. when a fdt induction algorithm utilizes stopping criteria for early stopping of the tree's growth, threshold values of stopping criteria will control the number of nodes. finding a proper threshold value for a stopping crite...

In this paper we briefly study the basic idea of Akaike's (1973) information criterion (AIC). Then, we present some recent developments on a new entropic or information complexity (ICOMP) criterion of Bozdogan (1988a, 1988b, 1990, 1994d, 1996, 1998a, 1998b) for model selection. A rationale for ICOMP as a model selection criterion is that it combines a badness-of-fit term (such as minus twice th...

Factor analysis is a statistical procedure to describe observed data in terms of unobserved variables called factors. Naturally, it is necessary to determine the number of factors to represent the system. There are several existent criteria to deal with the tradeoff between reduction of approximation error and avoidance of overparameterization. However, given the factors there is a lack of an a...

For a given source distribution, we establish properties of the conditional density achieving the rate distortion function lower bound as the distortion parameter varies. In the limit as the distortion tolerated goes to zero, the conditional density achieving the rate distortion function lower bound becomes degenerate in the sense that the channel it defines becomes error-free. As the permitted...

Using an innovations state space approach, it has been found that the Akaike information criterion (AIC) works slightly better, on average, than prediction validation on withheld data, for choosing between the various common methods of exponential smoothing for forecasting. There is, however, a puzzle. Should the count of the seed states be incorporated into the penalty term in the AIC formula?...

In failure time distributions, inliers in a data set are subset of observations sufficiently small relative to the rest of the observations, which appears to be inconsistent with the remaining data set. They are either the resultant of instantaneous failures or early failures, experienced in many life-testing experiments. The model used in outliers, where r observations are outliers is modified...

We extend the basic idea of Schwarz (1978) and develop a generalized Bayesian information criterion for regression model selection. The new criterion relaxes the usually strong distributional assumption associated with Schwarz’s BIC by adopting a Wilcoxon-type dispersion function and appropriately adjusting the penalty term. We establish that the Wilcoxon-type generalized BIC preserves the cons...

We know that I(̺,M) and uncertainty δ2(M)̺ coincides on pure state, i.e, when ̺ is a pure state, then the inequalities (2) and (6) in lemma 1, 2 both become equality. let us recall the “Local Uncertainty Relations” (LURs), introduced by Hofmann and Takeuchi [4]. Let Ai be observables on Alice’s space of a bipartite system. If they do not share a common eigenstate, there is a number CA > 0 such tha...

Evaluating the generalization performance of learning machines without using additional test samples is one of the most important issues in the machine learning community. The subspace information criterion (SIC) is one of the methods for this purpose, which is shown to be an unbiased estimator of the generalization error with finite samples. Although the mean of SIC agrees with the true genera...