CoChat
Sign Up
Cited by 51,879Statistical Methods and InferenceControl Systems and IdentificationStatistical and numerical algorithms

A new look at the statistical model identification

Akaike, H.·1974·IEEE Transactions on Automatic Control
DOI 10.1109/tac.1974.1100705
Open this paper in CoChat — free account

Abstract

The history of the development of statistical hypothesis testing in time series analysis is reviewed briefly and it is pointed out that the hypothesis testing procedure is not adequately defined as the procedure for statistical model identification. The classical maximum likelihood estimation procedure is reviewed and a new estimate minimum information theoretical criterion (AIC) estimate (MAICE) which is designed for the purpose of statistical identification is introduced. When there are several competing models the MAICE is defined by the model and the maximum likelihood estimates of the parameters which give the minimum of AIC defined by AIC = (-2)log-(maximum likelihood) + 2(number of independently adjusted parameters within the model). MAICE provides a versatile procedure for statistical model identification which is free from the ambiguities inherent in the application of conventional hypothesis testing procedure. The practical utility of MAICE in time series analysis is demonstrated with some numerical examples.

Integrity signals

Registry

Crossref · 10.1109/tac.1974.1100705

Cites

47 of 47 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AIC and ECVI can be used to compare competing models that need not be nested, with smaller values indicating better fit.

    “Within a set of models for the same data, the Akaike Information Criterion (AIC; Akaike, 1974) and Expected Cross Validation Index (ECVI; Browne & Cudeck, 1993) can be used to compare competing models that need not be nested (smaller is better).”

    From Nostalgia-Evoked Inspiration: Mediating Mechanisms and Motivational Implications (2015) · cites this paper as Akaike, 1974

  • The appropriateness of using OSMs with linear models (and linear mixed models) was evaluated by computing AIC and BIC of different models for attractiveness and each of the four dependent variables.

    “In addition to the theoretical grounding, we further evaluated the appropriateness of using the OSMs with linear models (and linear mixed models) by computing the Akaike information criterion (AIC) [120] and the Bayesian information criterion (BIC) [121] of different models for attractiveness and each of the four dependent variables.”

    From What is beautiful is still good: the attractiveness halo effect in the era of beauty filters (2024) · cites this paper as [120]

  • The quality of the statistical model was checked using the Akaike information criterion (AIC).

    “The quality of the statistical model was checked using the entropy maximization principle of the Akaike information criterion (AIC) (Akaike, 1974).”

    From Persistent psychopathology and neurocognitive impairment in COVID-19 survivors: effect of inflammatory biomarkers at three-month follow-up (2021) · cites this paper as Akaike, 1974

  • Model fits were compared using BIC and AIC, which penalise models for the number of free parameters to prevent overfitting.

    “Model fits were compared using the BIC and the AIC [34], which penalise models for the number of free parameters that are estimated to ensure that models with more parameters are not overfitted.”

    From Neurocomputational mechanisms underlying subjective valuation of effort costs (2017) · cites this paper as [34]

  • The best model was identified using Akaike's (AIC) and Schwartz's information criteria (SBC or BIC).

    “The best model was identified using Akaike's (AIC) and Schwartz's information criteria (SBC or BIC) (45,46).”

    From Influence of the COMT genotype on working memory and brain activity changes during development (2011) · cites this paper as [45]

Cites

Flagged first, then by citations

1 of the 48 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Akaike, H.
Affiliations
The Institute of Statistical Mathematics (JP)
Corresponding author
Akaike, H.
Venue
IEEE Transactions on Automatic Control
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Published
1974-12
Volume
19
Issue
6
Pages
716-723
ISSN
0018-9286 (ISSN-L)
DOI
10.1109/tac.1974.1100705
OpenAlex
W2142635246
Cited by
51,879
References
48
Citations by year
2026: 2363 · 2025: 2646 · 2024: 2436 · 2023: 2545 · 2022: 2756 · 2021: 3085 · 2020: 2796 · 2019: 2643 · 2018: 2462 · 2017: 2474 · 2016: 2272 · 2015: 2369 · 2014: 2357 · 2013: 2274 · 2012: 2097
FWCI
42.49
Type
journal article
Language
en
Open access
closed
Licence
ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html (vor)
Locations
IEEE Transactions on Automatic Control (publishedVersion)
Topics
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
  • Physical Sciences › Engineering › Control and Systems Engineering › Control Systems and Identification
  • Physical Sciences › Mathematics › Applied Mathematics › Statistical and numerical algorithms
Keywords
time series analysis, hypothesis testing, maximum likelihood estimation, Akaike information criterion, model selection

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review