Small-sample one-sided testing in extreme value regression models

Variable Selection in Regression Models with One-sided Information and a Small Sample

Given that one-sided hypothesis tests are more powerful than their two-sided counterparts, model selection procedures that employ one-sided information should outperform those that do not. In the frequentist area of econometrics, perhaps the most popular model selection procedures are those based on estimates of the Kullback-Leibler information, Akaike’s (1973) Information Criterion (AIC) being...

Testing for one-sided alternatives in nonparametric cen- sored regression

Assume we have two populations satisfying the general model Yj = mj(Xj)+εj , j = 1, 2, where m(·) is a smooth function, ε has zero location and Yj is possibly right-censored. In this paper, we propose to test the null hypothesis H0 : m1 = m2 versus the one-sided alternative H1 : m1 < m2. We introduce two test statistics for which we obtain the asymptotic normality under the null and the alterna...

Small-sample Forecasting Regression or Arima Models?

Extreme Value Control Charts in Some Life Testing Models

Variable control charts are based on sub group statistics and variation in the values of the sub group statistics between sub groups. In this paper extreme order statistic of the sub group are considered to develop to the control limits to decide upon the in control status of the process. The quality characteristic is assumed to follow Exponential/Weibull Distributions. Relevant comparisons are...

Estimating bank default with generalised extreme value regression models

This paper proposes a novel model for the prediction of bank failures, on the basis of both macroeconomic and bank-specific microeconomic factors. As bank failures are rare, in the paper we apply a regression method based on extreme value theory, which turns out to be more effective than classical logistic regression models, as it better leverages the information in the tail of the default dist...