Abstract
We present a multivariate sufficient statistic using Kronecker products that dramatically increases computational efficiency in evaluating likelihood functions and/or posterior distributions. In particular, we examine the example of multivariate regression in a Bayesian setting. We compare the computation time for using the Gibbs sampler both with and without the sufficient statistic. We show that the difference in computation time grows quadratically with the number of covariates and products and linearly with the number of individuals. In the simulation study, speedup factors ranging from at least 20 times to almost 300 times were observed when using the Kronecker sufficient statistic.
Similar content being viewed by others
References
Barnett, S.: Matrix differential equations and Kronecker products. SIAM J. Appl. Math. 24(1), 1–5 (1973)
Casella, G., Berger, R.L.: Statistical Inference. Duxbury Press, Belmont (1990)
Davidson, R.R., Beaver, R.J.: On extending the Bradley-Terry model to incorporate within-pair order effects. Biometrics 33, 693–702 (1977)
Fisher, W.D.: Optimal aggregation in multi-equation prediction models. Econometrica 30(4), 744–769 (1962)
Gelfand, A.E., Smith, A.F.M.: Sampling-based approaches to calculating marginal densities. J. Am. Stat. Assoc. 85, 398–409 (1990)
Hultquist, R.A., Graybill, F.A.: Minimal sufficient statistics for the two-way classification mixed model design. J. Am. Stat. Assoc. 60(309), 182–192 (1965)
Searle, S.R.: Matrix Algebra Useful for Statistics, 2nd edn. Wiley, New York (2006). 265 & 333
Van Loan, C.: The ubiquitous Kronecker product. J. Comput. Appl. Math. 123, 85–100 (2000)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liechty, M.W., Tibbits, M. Multivariate sufficient statistics using Kronecker products. Stat Comput 20, 335–341 (2010). https://doi.org/10.1007/s11222-009-9127-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-009-9127-x