Empirical Bayes methods are procedures for statistical inference in which the prior distribution is estimated from the data. This approach stands in contrast to standard Bayesian methods, for which the prior distribution is fixed before any data are observed. Despite this difference in perspective, empirical Bayes may be viewed as an approximation to a fully Bayesian treatment of a hierarchical model wherein the parameters at the highest level of the hierarchy are set to their most likely values, instead of being integrated out. Empirical Bayes, also known as maximum marginal likelihood,^{[1]} represents one approach for setting hyperparameters.
Contents

Introduction 1

Point estimation 2

Robbins method : nonparametric empirical Bayes (NPEB) 2.1

Parametric empirical Bayes 2.2

Poisson–gamma model 2.2.1

See also 3

References 4

Further reading 5

External links 6
Introduction
Empirical Bayes methods can be seen as an approximation to a fully Bayesian treatment of a hierarchical Bayes model.
In, for example, a twostage hierarchical Bayes model, observed data y = \{y_1, y_2, \dots, y_n\} are assumed to be generated from an unobserved set of parameters \theta = \{\theta_1, \theta_2, \dots, \theta_n\} according to a probability distribution p(y\theta)\,. In turn, the parameters \theta can be considered samples drawn from a population characterised by hyperparameters \eta\, according to a probability distribution p(\theta\eta)\,. In the hierarchical Bayes model, though not in the empirical Bayes approximation, the hyperparameters \eta\, are considered to be drawn from an unparameterized distribution p(\eta)\,.
Information about a particular quantity of interest \theta_i\; therefore comes not only from the properties of those data which directly depend on it, but also from the properties of the population of parameters \theta\; as a whole, inferred from the data as a whole, summarised by the hyperparameters \eta\;.
Using Bayes' theorem,

p(\thetay) = \frac{p(y  \theta) p(\theta)}{p(y)} = \frac {p(y  \theta)}{p(y)} \int p(\theta  \eta) p(\eta) \, d\eta \,.
In general, this integral will not be tractable analytically or symbolically and must be evaluated by numerical methods. Stochastic (random) or deterministic approximations may be used. Example stochastic methods are Markov Chain Monte Carlo and Monte Carlo sampling. Deterministic approximations are discussed in quadrature.
Alternatively, the expression can be written as

p(\thetay) = \int p(\theta\eta, y) p(\eta  y) \; d \eta = \int \frac{p(y  \theta) p(\theta  \eta)}{p(y  \eta)} p(\eta  y) \; d \eta\,,
and the term in the integral can in turn be expressed as

p(\eta  y) = \int p(\eta  \theta) p(\theta  y) \; d \theta .
These suggest an iterative scheme, qualitatively similar in structure to a Gibbs sampler, to evolve successively improved approximations to p(\thetay)\; and p(\etay)\;. First, calculate an initial approximation to p(\thetay)\; ignoring the \eta dependence completely; then calculate an approximation to p(\etay)\; based upon the initial approximate distribution of p(\thetay)\;; then use this p(\etay)\; to update the approximation for p(\thetay)\;; then update p(\etay)\;; and so on.
When the true distribution p(\etay)\; is sharply peaked, the integral determining p(\thetay)\; may be not much changed by replacing the probability distribution over \eta\; with a point estimate \eta^{*}\; representing the distribution's peak (or, alternatively, its mean),

p(\thetay) \simeq \frac{p(y  \theta) \; p(\theta  \eta^{*})}{p(y  \eta^{*})}\,.
With this approximation, the above iterative scheme becomes the EM algorithm.
The term "Empirical Bayes" can cover a wide variety of methods, but most can be regarded as an early truncation of either the above scheme or something quite like it. Point estimates, rather than the whole distribution, are typically used for the parameter(s) \eta\;. The estimates for \eta^{*}\; are typically made from the first approximation to p(\thetay)\; without subsequent refinement. These estimates for \eta^{*}\; are usually made without considering an appropriate prior distribution for \eta.
Point estimation
Robbins method : nonparametric empirical Bayes (NPEB)
Robbins^{[2]} considered a case of sampling from a mixed distribution, where probability for each y_i (conditional on \theta_i) is specified by a Poisson distribution,

p(y_i\theta_i)=,
where p_{G} is the marginal distribution obtained by integrating out θ over G.
To take advantage of this, Robbins^{[2]} suggested estimating the marginals with their empirical frequencies, yielding the fully nonparametric estimate as:

\operatorname{E}(\theta_iy_i) \approx (y_i + 1) { {\#\{Y_j = y_i + 1\}} \over {\#\{ Y_j = y_i\}} },
where \# denotes "number of". (See also Good–Turing frequency estimation.)

Example  Accident rates
Suppose each customer of an insurance company has an "accident rate" Θ and is insured against accidents; the probability distribution of Θ is the underlying distribution, and is unknown. The number of accidents suffered by each customer in a specified time period has a Poisson distribution with expected value equal to the particular customer's accident rate. The actual number of accidents experienced by a customer is the observable quantity. A crude way to estimate the underlying probability distribution of the accident rate Θ is to estimate the proportion of members of the whole population suffering 0, 1, 2, 3, ... accidents during the specified time period as the corresponding proportion in the observed random sample. Having done so, it is then desired to predict the accident rate of each customer in the sample. As above, one may use the conditional expected value of the accident rate Θ given the observed number of accidents during the baseline period. Thus, if a customer suffers six accidents during the baseline period, that customer's estimated accident rate is 7 × [the proportion of the sample who suffered 7 accidents] / [the proportion of the sample who suffered 6 accidents]. Note that if the proportion of people suffering k accidents is a decreasing function of k, the customer's predicted accident rate will often be lower than their observed number of accidents. This shrinkage effect is typical of empirical Bayes analyses.
Parametric empirical Bayes
If the likelihood and its prior take on simple parametric forms (such as 1 or 2dimensional likelihood functions with simple conjugate priors), then the empirical Bayes problem is only to estimate the marginal m(y\eta) and the hyperparameters \eta using the complete set of empirical measurements. For example, one common approach, called parametric empirical Bayes point estimation, is to approximate the marginal using the maximum likelihood estimate (MLE), or a Moments expansion, which allows one to express the hyperparameters \eta in terms of the empirical mean and variance. This simplified marginal allows one to plug in the empirical averages into a point estimate for the prior \theta. The resulting equation for the prior \theta is greatly simplified, as shown below.
There are several common parametric empirical Bayes models, including the Poisson–gamma model (below), the Betabinomial model, the Gaussian–Gaussian model, the Dirichletmultinomial model, as well specific models for Bayesian linear regression (see below) and Bayesian multivariate linear regression. More advanced approaches include hierarchical Bayes models and Bayesian mixture models.
Poisson–gamma model
For example, in the example above, let the likelihood be a Poisson distribution, and let the prior now be specified by the conjugate prior, which is a gamma distribution (G(\alpha,\beta)) (where \eta = (\alpha,\beta)):

\rho(\theta\alpha,\beta) = \frac{\theta^{\alpha1}\, e^{\theta / \beta} }{\beta^{\alpha} \Gamma(\alpha)} \ \mathrm{for}\ \theta > 0, \alpha > 0, \beta > 0 \,\! .
It is straightforward to show the posterior is also a gamma distribution. Write

\rho(\thetay) \propto \rho(y\theta) \rho(\theta\alpha, \beta) ,
where the marginal distribution has been omitted since it does not depend explicitly on \theta. Expanding terms which do depend on \theta gives the posterior as:

\rho(\thetay) \propto (\theta^{y}\, e^{\theta}) (\theta^{\alpha1}\, e^{\theta / \beta}) = \theta^{y+ \alpha 1}\, e^{ \theta (1+1 / \beta)} .
So the posterior density is also a gamma distribution G(\alpha',\beta'), where \alpha' = y + \alpha, and \beta' = (1+1 / \beta)^{1}. Also notice that the marginal is simply the integral of the posterior over all \Theta, which turns out to be a negative binomial distribution.
To apply empirical Bayes, we will approximate the marginal using the maximum likelihood estimate (MLE). But since the posterior is a gamma distribution, the MLE of the marginal turns out to be just the mean of the posterior, which is the point estimate \operatorname{E}(\thetay) we need. Recalling that the mean \mu of a gamma distribution G(\alpha', \beta') is simply \alpha' \beta', we have

\operatorname{E}(\thetay) = \alpha' \beta' = \frac{\bar{y}+\alpha}{1+1 / \beta} = \frac{\beta}{1+\beta}\bar{y} + \frac{1}{1+\beta} (\alpha \beta).
To obtain the values of \alpha and \beta, empirical Bayes prescribes estimating mean \alpha\beta and variance \alpha\beta^2 using the complete set of empirical data.
The resulting point estimate \operatorname{E}(\thetay) is therefore like a weighted average of the sample mean \bar{y} and the prior mean \mu = \alpha\beta. This turns out to be a general feature of empirical Bayes; the point estimates for the prior (i.e. mean) will look like a weighted averages of the sample estimate and the prior estimate (likewise for estimates of the variance).
See also
References

^ C.M. Bishop (2005). Neural networks for pattern recognition. Oxford University Press ISBN 0198538642

^ ^{a} ^{b}

^ Carlin, Bradley P.; Louis, Thomas A. (2000). Bayes and Empirical Bayes Methods for Data Analysis (2nd ed.). Chapman & Hall/CRC. pp. Sec. 3.2 and Appendix B.
Further reading

Peter E. Rossi, Greg M. Allenby, and Robert McCulloch, Bayesian Statistics and Marketing, John Wiley & Sons, Ltd, 2006

Casella, George (May 1985). "An Introduction to Empirical Bayes Data Analysis". American Statistician (American Statistical Association) 39 (2): 83–87.

Nikulin, Mikhail (1987). "Bernstein's regularity conditions in a problem of empirical Bayesian approach". Journal of Soviet Mathematics, 36 (5): 596–600.
External links

Use of empirical Bayes Method in estimating road safety (North America)

Empirical Bayes methods for missing data analysis

Using the BetaBinomial distribution to assess performance of a biometric identification device

A Hierarchical Naive Bayes Classifiers (for continuous and discrete variables).
This article was sourced from Creative Commons AttributionShareAlike License; additional terms may apply. World Heritage Encyclopedia content is assembled from numerous content providers, Open Access Publishing, and in compliance with The Fair Access to Science and Technology Research Act (FASTR), Wikimedia Foundation, Inc., Public Library of Science, The Encyclopedia of Life, Open Book Publishers (OBP), PubMed, U.S. National Library of Medicine, National Center for Biotechnology Information, U.S. National Library of Medicine, National Institutes of Health (NIH), U.S. Department of Health & Human Services, and USA.gov, which sources content from all federal, state, local, tribal, and territorial government publication portals (.gov, .mil, .edu). Funding for USA.gov and content contributors is made possible from the U.S. Congress, EGovernment Act of 2002.
Crowd sourced content that is contributed to World Heritage Encyclopedia is peer reviewed and edited by our editorial staff to ensure quality scholarly research articles.
By using this site, you agree to the Terms of Use and Privacy Policy. World Heritage Encyclopedia™ is a registered trademark of the World Public Library Association, a nonprofit organization.