Regression and time series model selection in small samples

1989/01/01 by Clifford M. Hurvich, Chih‐Ling Tsai · 35 citations
Engineering · Mathematics · Computer Science · #Control Systems and Identification #Statistical Methods and Inference #Neural Networks and Applications #Akaike information criterion #Autoregressive model #Mathematics #STAR model #Series (stratigraphy) #Model selection #SETAR #Statistics #Sample size determination #Autoregressive integrated moving average #Information Criteria #Applied mathematics #Time series #Dimension (graph theory) #Regression analysis #Selection (genetic algorithm) #Combinatorics #Computer science #Artificial intelligence

paper · doi:10.1093/biomet/76.2.297

openalex publication_date 1989/01/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

A bias correction to the Akaike information criterion, AIC, is derived for regression and autoregressive time series models. The correction is of particular use when the sample size is small, or when the number of fitted parameters is a moderate to large fraction of the sample size. The corrected method, called AICC, is asymptotically efficient if the true model is infinite dimensional. Furthermore, when the true model is of finite dimension, AICC is found to provide better model order choices than any other asymptotically efficient method. Applications to nonstationary autoregressive and mixed autoregressive moving average time series models are also discussed.

Citations

• The Estimation of the Order of an ARMA Process
• The Relationship Between Variable Selection and Data Agumentation and a Method for Prediction

Cited by

• Model Selection
• Regression model selection via log‐likelihood ratio and constrained minimum criterion
• The effect of protein timing on muscle strength and hypertrophy: a meta-analysis
• Adaptive Nonlinear Data Assimilation through P-Spline Triangular Measure Transport
• Phase-space networks and connectivity of the kagome antiferromagnet
• Unifying the derivations for the Akaike and corrected Akaike information criteria
• Bridging AIC and BIC: A New Criterion for Autoregression
• Evolution of multicellularity coincided with increased diversification of cyanobacteria and the Great Oxidation Event
• Combining time series models for forecasting
• Resource orchestration of firm-specific human capital and firm performance—the role of collaborative human resource management and entrepreneurial orientation
• Kullback-Leibler information as a basis for strong inference in ecological studies
• Lemurs in a dying forest: Factors influencing lemur diversity and distribution in forest remnants of north-eastern Madagascar
• Mammuthus Population Dynamics in Late Pleistocene North America: Divergence, Phylogeography, and Introgression
• HSTPROMO Internal Proper-motion Kinematics of Dwarf Spheroidal Galaxies. I. Velocity Anisotropy and Dark Matter Cusp Slope of Draco
• Extending Statistical Boosting
• A predictive analytics approach to reducing avoidable hospital readmission
• Minimal Penalties for Gaussian Model Selection
• Discrete Bilal distribution with right-censored data
• Fast Uplink Grant-Free NOMA with Sinusoidal Spreading Sequences
• Selection of tuning parameters in bridge regression models via Bayesian information criterion
• THE MASS OF Kepler-93b AND THE COMPOSITION OF TERRESTRIAL PLANETS
• Evaluating the New Automatic Method for the Analysis of Absorption Spectra Using Synthetic Spectra
• BODY MASS AND FORAGING ECOLOGY PREDICT EVOLUTIONARY PATTERNS OF SKELETAL PNEUMATICITY IN THE DIVERSE “WATERBIRD” CLADE
• ON THE CLASSIFICATION OF UGC 1382 AS A GIANT LOW SURFACE BRIGHTNESS GALAXY
• Information-based inference for singular models and finite sample sizes: A frequentist information criterion
• Parametrization of microbial survival models under UVC exposure
• Non-uniqueness in quasar absorption models and implications for measurements of the fine structure constant
• Combining predictions from linear models when training and test inputs\n differ
• Reverse migration as a cause of vagrancy
• Large wood as a confounding factor in interpreting the width of spring-fed streams
• Artificial intelligence applied to the automatic analysis of absorption spectra. Objective measurement of the fine structure constant.
• A NON-PARAMETRIC APPROACH TO CONSTRAIN THE TRANSFER FUNCTION IN REVERBERATION MAPPING
• Determination of the charm quark mass in lattice QCD with 2 + 1 flavours on fine lattices
• Factors affecting the detection probability of a critically endangered flying-fox: consequences for monitoring and conservation
• Akaike information criterion [wikipedia]