AIC, Overfitting Principles, and the Boundedness of Moments of Inverse Matrices for Vector Autotregressions and Related Models

@article{Findley2002AICOP,
  title={AIC, Overfitting Principles, and the Boundedness of Moments of Inverse Matrices for Vector Autotregressions and Related Models},
  author={David F. Findley and Ching-Zong Wei},
  journal={Journal of Multivariate Analysis},
  year={2002},
  volume={83},
  pages={415-450}
}
In his somewhat informal derivation, Akaike (in "Proceedings of the 2nd International Symposium Information Theory" (C. B. Petrov and F. Csaki, Eds.), pp. 610-624, Academici Kiado, Budapest, 1973) obtained AIC's parameter-count adjustment to the log-likelihood as a bias correction: it yields an asymptotically unbiased estimate of the quantity that measures the average fit of the estimated model to an independent replicate of the data used for estimation. We present the first mathematically… 
Efficient and Consistent Data-Driven Model Selection for Time Series
TLDR
This paper proves that consistent model selection criteria outperform classical AIC criterion in terms of efficiency and derives from a Bayesian approach the usual BIC criterion, by keeping all the second order terms of the Laplace approximation, a data-driven criterion denoted KC’.
Order selection for same-realization predictions in autoregressive processes
TLDR
This paper presents the first theoretical verification that AIC and its variants are still asymptotically elficient (in the sense defined in Section 4) for same-realization predictions, and shows that A IC also yields a satisfactory saute- realization prediction in finite samples.
Statistica Sinica Preprint No : SS-2016-0069 . R 2 Title Nearly Unstable Processes : A Prediction Perspective
Prediction has long been a vibrant topic in modern probability and statistics. In addition to finding optimal forecast and model selection, it is argued in this paper that the prediction principle
Nearly Unstable Processes: A Prediction Perspective
Prediction has long been a vibrant topic in modern probability and statistics. In addition to finding optimal forecast and model selection, it is argued in this paper that the prediction principle
Forecast Averaging with Panel Data Vector Autoregressions
In this paper we propose a new forecast model averaging method for panel data vector autoregressions that permit limited forms of parameterized heterogeneity (including fixed effects or incidental
Moment bounds and mean squared prediction errors of long-memory time series
A moment bound for the normalized conditional-sum-of-squares (CSS) estimate of a general autoregressive fractionally integrated moving average (ARFIMA) model with an arbitrary unknown memory
MULTISTEP PREDICTION OF PANEL VECTOR AUTOREGRESSIVE PROCESSES
This paper considers the conventional recursive (otherwise known as plug-in) and direct multistep forecasts in a panel vector autoregressive framework. We derive asymptotic expressions for the mean
Information Criterion for Minimum Cross-Entropy Model Selection
TLDR
An iterative method to approximate the target density by minimizing the cross-entropy information criterion (CIC) is proposed and it is demonstrated that the proposed method selects a parametric model that well approximates the targetdensity.
FROM A FINITE FAMILY OF POSSIBLY MISSPECIFIED TIME SERIES MODELS By
TLDR
A misspecification-resistant information criterion (MRIC) is proposed and it is shown that MRIC can be used in conjunction with a high-dimensional model selection method to select the (asymptotically) best predictive model across several high- dimensional misspecified time series models.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 35 REFERENCES
ON THE UNBIASEDNESS PROPERTY OF AIC FOR EXACT OR APPROXIMATING LINEAR STOCHASTIC TIME SERIES MODELS
. A rigorous analysis is given of the asymptotic bias of the log maximum likelihood as an estimate of the expected log likelihood of the maximum likelihood model, when a linear model, such as an
Counterexamples to parsimony and BIC
Suppose that the log-likelihood-ratio sequence of two models with different numbers of estimated parameters is bounded in probability, without necessarily having a chi-square limiting distribution.
PREDICTION ERROR OF MULTIVARIATE TIME SERIES WITH MIS-SPECIFIED MODELS
We consider the situation in which an incorrectly specified autoregressive moving-average model is used to predict future values of a stationary multivariate time series. The use of an incorrect
Model Selection Tests for Nonlinear Dynamic Models
This paper generalizes Vuong (1989) asymptotically normal tests for model selection in several important directions. First, it allows for incompletely parametrized models such as econometric models
Introduction to Akaike (1973) Information Theory and an Extension of the Maximum Likelihood Principle
The problem of estimating the dimensionality of a model occurs in various forms in applied statistics: estimating the number of factors in factor analysis, estimating the degree of a polynomial
Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
Using the Kullback-Leibler information criterion to measure the closeness of a model to the truth, the author proposes new likelihood-ratio-based statistics for testing the null hypothesis that the
Convergence analysis of parametric identification methods
  • L. Ljung
  • Computer Science, Mathematics
  • 1978
A certain class of methods to select suitable models of dynamical stochastic systems from measured input-output data is considered. The methods are based on a comparison between the measured outputs
Subsampling and model selection in time series analysis
In this paper the subsampling method of Carlstein (1986) is used to estimate the risk of prediction for time series data. We extend Carlstein's result by proving strong consistency of the subsampling
Asymptotic Inference about Predictive Ability
This paper develops procedures for inference about the moments of smooth functions of out-of-sample predictions and prediction errors when there is a long time series of predictions and realizations.
Maximum likelihood estimates of incorrect Markov models for time series and the derivation of AIC
The asymptotic behavior of the maximum likelihood estimators of Markov models or autoregressive models are given when the true distribution is not a member of the assumed parametric family. The
...
1
2
3
4
...