Statistical predictor identification

  title={Statistical predictor identification},
  author={Hirotugu Akaike},
  journal={Annals of the Institute of Statistical Mathematics},
  • H. Akaike
  • Published 1 December 1970
  • Mathematics
  • Annals of the Institute of Statistical Mathematics
In a recent paper by the present author [1] a simple practical procedure of predictor identification has been proposed. It is the purpose of this paper to provide a theoretical and empirical basis of the procedure. 
Abstract. A procedure based on the automatic information criterion procedure of Akaike is presented for estimating the number of sinusoidal terms present in a time series. The procedure is shown to
. Formulae are obtained for the limiting probability of overestimating the order of a multivariate autoregression when using AIC-type procedures.
On the Use of AIC for the Detection of Outliers
An objective procedure for the detection of outliers is given by using Akaike's information criterion. Numerical illustrations are given, using data from Grubbs [8] and Tietjen and Moore [7].
An optimal selection of regression variables
SUMMARY An asymptotically optimal selection of regression variables is proposed. The key assumption is that the number of control variables is infinite or increases with the sample size. It is also
On a criterion for the selection of models for stationary time series
SummaryA general approach to model selection is described and a criterion for the selection of parametric models for stationary Gaussian time series is derived. A corresponding approximate test is
A survey of time series analysis through parametric models
The paper is a survey of recent works on time series analysis using parametric models. The main emphasis is on linear models, in particular the ARMA model. Usual me¬thods of parameter estimation,
The Levinson Algorithm and its Applications in Time Series Analysis
Summary This paper presents the Levinson algorithm, its rediscoveries and extensions. A review is given of the applications of the procedure to several areas of interest in the field of time series
Akaike’s Approach can Yield Consistent Order Determination
More than twenty years ago, Akaike proposed what is nowadays called the AIC approach to the determination of the order of a linear autoregressive (LAR) model.
Order Determination for a Multivariate Autoregression
SUMMARY The procedure proposed by Hannan and Quinn (1979) for determining the order of univariate autoregressions is extended to multivariate autoregressions. The procedure is shown to be strongly
Effects of Model Selection on Inference
The asymptotic properties of parameter estimators which are based on a model that has been selected by a model selection procedure are investigated. In particular, the asymptotic distribution is


Prediction of Multivariate Time Series
Abstract A method of linear prediction of stationary multivariate time series is discussed from the point of view of meteorological applications. Tests of significance are given and it is shown by
Fitting autoregressive models for prediction
This is a preliminary report on a newly developed simple and practical procedure of statistical identification of predictors by using autoregressive models. The use of autoregressive representation
Some probability limit theorems with statistical applications
In fundamental papers Bernstein (3) and Loeve(8) have proved central limit theorems for wide classes of dependent variables. Their theorems are stated in terms of conditional distributions. In the