Akaike information criterion

From Wikipedia, the free encyclopedia

Jump to: navigation, search
The Akaike information criterion (AIC) is a measure of the relative quality of a statistical
model, for a given set of data. As such, AIC provides a means for model selection.
AIC deals with the trade-off between the goodness of fit of the model and the complexity of
the model. It is founded on information entropy: it offers a relative estimate of the
information lost when a given model is used to represent the process that generates the data.
AIC does not provide a test of a model in the sense of testing a null hypothesis; i.e. AIC can
tell nothing about the quality of the model in an absolute sense. If all the candidate models fit
poorly, AIC will not give any warning of that.
Contents
[hide]
1 Definition
2 How to apply AIC in practice
3 AICc
4 Relevance to chi-squared fitting
5 History
6 Comparison with BIC
7 See also
8 Notes
9 References
10 External links
Definition[edit]
In the general case, the AIC is

where k is the number of parameters in the statistical model, and L is the maximized value of
the likelihood function for the estimated model.
Given a set of candidate models for the data, the preferred model is the one with the minimum
AIC value. Hence AIC not only rewards goodness of fit, but also includes a penalty that is an
increasing function of the number of estimated parameters. This penalty discourages
overfitting (increasing the number of free parameters in the model improves the goodness of
the fit, regardless of the number of free parameters in the data-generating process).
AIC is founded in information theory. Suppose that the data is generated by some unknown
process f. We consider two candidate models to represent f: g
1
and g
2
. If we knew f, then we
could find the information lost from using g
1
to represent f by calculating the Kullback
Leibler divergence, D
KL
(f g
1
); similarly, the information lost from using g
2
to represent f
would be found by calculating D
KL
(f g
2
). We would then choose the candidate model that
minimized the information loss.
We cannot choose with certainty, because we do not know f. Akaike (1974) showed,
however, that we can estimate, via AIC, how much more (or less) information is lost by g
1

than by g
2
. It is remarkable that such a simple formula for AIC results. The estimate, though,
is only valid asymptotically; if the number of data points is small, then some correction is
often necessary (see AICc, below).
How to apply AIC in practice[edit]
To apply AIC in practice, we start with a set of candidate models, and then find the models'
corresponding AIC values. There will almost always be information lost due to using one of
the candidate models to represent the "true" model (i.e. the process that generates the data).
We wish to select, from among R candidate models, the model that minimizes the
information loss. We cannot choose with certainty, but we can minimize the estimated
information loss.
Denote the AIC values of the candidate models by AIC
1
, AIC
2
, AIC
3
, , AIC
R
. Let AIC
min

be the minimum of those values. Then exp((AIC
min
AIC
i
)/2) can be interpreted as the relative
probability that the ith model minimizes the (estimated) information loss.
[1]

As an example, suppose that there were three models in the candidate set, with AIC values
100, 102, and 110. Then the second model is exp((100102)/2) = 0.368 times as probable as
the first model to minimize the information loss, and the third model is exp((100110)/2) =
0.007 times as probable as the first model to minimize the information loss.
In this example, we would omit the third model from further consideration. We then have
three options: (1) we could decide to gather more data, in the hope that this will allow clearly
distinguishing between the first two models; (2) we could simply conclude that the data is
insufficient to support selecting one model from among the first two; (3) we could take a
weighted average of the first two models, with weights 1 and 0.368, respectively, and then do
statistical inference based on the weighted multimodel.
[2]

The quantity exp((AIC
min
AIC
i
)/2) is the relative likelihood of model i.
If all the models in the candidate set have the same number of parameters, then using AIC
might at first appear to be very similar to using the likelihood-ratio test. There are, however,
important distinctions. In particular, the likelihood-ratio test is valid only for nested models
whereas AIC (and AICc) has no such restriction.
[3]

AICc[edit]
AICc is AIC with a correction for finite sample sizes:

where n denotes the sample size. Thus, AICc is AIC with a greater penalty for extra
parameters.
Burnham & Anderson (2002) strongly recommend using AICc, rather than AIC, if n is small
or k is large. Since AICc converges to AIC as n gets large, AICc generally should be
employed regardless.
[4]
Using AIC, instead of AICc, when n is not many times larger than k
2
,
increases the probability of selecting models that have too many parameters, i.e. of
overfitting. The probability of AIC overfitting can be substantial, in some cases.
[5]

Brockwell & Davis (1991, p. 273) advise using AICc as the primary criterion in selecting the
orders of an ARMA model for time series. McQuarrie & Tsai (1998) ground their high
opinion of AICc on extensive simulation work with regression and time series.
AICc was first proposed by Hurvich & Tsai (1989). Different derivations of it are given by
Brockwell & Davis (1991), Burnham & Anderson, and Cavanaugh (1997). All the
derivations assume a univariate linear model with normally distributed errors (conditional
upon regressors); if that assumption does not hold, then the formula for AICc will usually
change. Further discussion of this, with examples of other assumptions, is given by Burnham
& Anderson (2002, ch. 7). In particular, bootstrap estimation is usually feasible.
Note that when all the models in the candidate set have the same k, then AICc and AIC will
give identical (relative) valuations. In that situation, then, AIC can always be used.
Relevance to chi-squared fitting[edit]
Often, one wishes to select amongst competing models where the likelihood functions
assume that the underlying errors are normally distributed (with mean zero) and independent.
This assumption leads to model fitting.
For fitting, the likelihood is given by


,
where C is a constant independent of the model used, and dependent only on the use of
particular data points. i.e. it does not change if the data do not change.
The AIC is therefore given by
. As only
differences in AIC are meaningful, the constant C can be ignored, allowing us to take
for model comparisons.
Another convenient form arises if the
i
are assumed to be identical and the residual sum of
squares (RSS) is available. Then we get AIC = n ln(RSS/n) + 2k + C, where again C can be
ignored in model comparisons.
[6]

History[edit]
The Akaike information criterion was developed by Hirotugu Akaike, under the name of "an
information criterion". It was first published by Akaike in 1974.
[7]

The original derivation of AIC relied upon some strong assumptions. Takeuchi (1976)
showed that the assumptions could be made much weaker. This work, however, was in
Japanese, and was not widely known outside Japan for many years.
AICc was originally proposed for linear regression (only) by Sugiura (1978). That instigated
the work of Hurvich & Tsai (1989), and several further papers by the same authors, which
extended the situations in which AICc could be applied. The work of Hurvich & Tsai
contributed to the decision to publish a second edition of the volume by Brockwell & Davis
(1991), which is the standard reference for linear time series; the new edition states, "our
prime criterion for model selection [among ARMA(p,q) models] will be the AICc".
[8]

The volume by Burnham & Anderson (2002) was the first attempt to set out the information-
theoretic approach in a general context. It includes an English exposition of the results of
Takeuchi. The volume led to far greater use of the information-theoretic approach, and now
has over 20000 citations on Google Scholar.
Akaike originally called his approach an entropy maximization principle. Burnham &
Anderson (2002, ch. 2) discuss and expand on this, and trace the approach back to the work
of Ludwig Boltzmann on thermodynamics. Briefly, minimizing AIC in a statistical model is
essentially equivalent to maximizing entropy in a thermodynamic system. In other words, the
information-theoretic approach in statistics is essentially applying the Second Law of
Thermodynamics.
Comparison with BIC[edit]
The AIC penalizes the number of parameters less strongly than does the Bayesian
information criterion (BIC). A comparison of AIC/AICc and BIC is given by Burnham &
Anderson (2002, 6.4). The authors show that AIC and AICc can be derived in the same
Bayesian framework as BIC, just by using a different prior. The authors also argue that
AIC/AICc has theoretical advantages over BIC. First, because AIC/AICc is derived from
principles of information; BIC is not, despite its name. Second, because the (Bayesian-
framework) derivation of BIC has a prior of 1/R (where R is the number of candidate
models), which is "not sensible", since the prior should be a decreasing function of k.
Additionally, they present a few simulation studies that suggest AICc tends to have
practical/performance advantages over BIC. See too Burnham & Anderson (2004).
Further comparison of AIC and BIC, in the context of regression, is given by Yang (2005). In
particular, AIC is asymptotically optimal in selecting the model with the least mean squared
error, under the assumption that the exact "true" model is not in the candidate set (as is
virtually always the case in practice); BIC is not asymptotically optimal under the
assumption. Yang further shows that the rate at which AIC converges to the optimum is, in a
certain sense, the best possible.
See also[edit]
Deviance information criterion
Focused information criterion
HannanQuinn information criterion
Occam's razor
Notes[edit]
1. Jump up ^ Burnham & Anderson 2002, 6.4.5
2. Jump up ^ Burnham & Anderson 2002
3. Jump up ^ Burnham & Anderson 2002, 2.12.4
4. Jump up ^ Burnham & Anderson 2004
5. Jump up ^ Claeskens & Hjort 2008, 8.3
6. Jump up ^ Burnham & Anderson 2002, pp. 12, 63
7. Jump up ^ Akaike 1974
8. Jump up ^ Brockwell & Davis 1991, p. 273
References[edit]
Akaike, Hirotugu (1974), "A new look at the statistical model identification", IEEE
Transactions on Automatic Control 19 (6): 716723,
doi:10.1109/TAC.1974.1100705, MR 0423716.
Akaike, Hirotugu (1980), "Likelihood and the Bayes procedure", in Bernardo, J. M.;
et al., Bayesian Statistics, Valencia: University Press, pp. 143166.
Anderson, D. R. (2008), Model Based Inference in the Life Sciences, Springer.
Brockwell, Peter J.; Davis, Richard A. (1987), Time Series: Theory and Methods,
Springer, ISBN 0387964061.
Brockwell, Peter J.; Davis, Richard A. (1991), Time Series: Theory and Methods (2nd
ed.), Springer, ISBN 0387974296. Republished in 2009: ISBN 1441903194
Burnham, K. P.; Anderson, D. R. (2002), Model Selection and Multimodel Inference:
A Practical Information-Theoretic Approach (2nd ed.), Springer-Verlag, ISBN 0-387-
95364-7.
Burnham, K. P.; Anderson, D. R. (2004), "Multimodel inference: understanding AIC
and BIC in Model Selection", Sociological Methods and Research 33: 261304.
Cavanaugh, J. E. (1997), "Unifying the derivations of the Akaike and corrected
Akaike information criteria", Statistics and Probability Letters 31: 201208.
Claeskens, G.; Hjort, N. L. (2008), Model Selection and Model Averaging,
Cambridge.
Fang, Yixin (2011). "Asymptotic equivalence between cross-validations and Akaike
Information Criteria in mixed-effects models", Journal of Data Science, 9:15-21.
Hurvich, C. M.; Tsai, C.-L. (1989), "Regression and time series model selection in
small samples", Biometrika 76: 297307.
Lukacs, P.M., et al. (2007). "Concerns regarding a call for pluralism of information
theory and hypothesis testing", Journal of Applied Ecology, 44:456460.
doi:10.1111/j.1365-2664.2006.01267.x.
McQuarrie, A. D. R.; Tsai, C.-L. (1998), Regression and Time Series Model
Selection, World Scientific, ISBN 981-02-3242-X.
Sugiura, N. (1978), "Further analysis of the data by Akaikes information criterion
and the finite corrections", Communications in Statistics - Theory and Methods A7:
1326.
Takeuchi, K. (1976), " " ["Distribution of informational statistics and a criterion of
model fitting"], Suri-Kagaku [Mathematical Sciences] (in Japanese) 153: 1218.
Yang, Y. (2005), "Can the strengths of AIC and BIC be shared?", Biometrika 92:
937950.
External links[edit]
Hirotogu Akaike comments on how he arrived at the AIC, in This Week's Citation
Classic (21 December 1981)
AIC (Aalto University)
Akaike Information Criterion (North Carolina State University)
Example AIC use (Honda USA, Noesis Solutions, Belgium)
Model Selection (University of Iowa)
<img src="//en.wikipedia.org/wiki/Special:CentralAutoLogin/start?type=1x1" alt="" title=""
width="1" height="1" style="border: none; position: absolute;" />
Retrieved from
"http://en.wikipedia.org/w/index.php?title=Akaike_information_criterion&oldid=584109897
"
Categories:
Regression variable selection
Model selection