Open Access Open Access  Restricted Access Subscription Access
Cover Image

Optimal design for count data

Muritala Abdulkabir, Udokang Anietie Edem, Bello Latifat Kemi

Abstract


Optimal designs for generalized linear models (GLM) have received increasing attention in recent years. Most of this research focuses on binary data model. This research extends to count data models. The aim and objectives of this research work to determine the appropriate generalized linear model (GLM) that is suitable for count data and identify a design that is best according to statistical optimality criteria, the data use for this research work are simulated data from R statistical package using uniform distribution with sample size 300. The simplest distribution use for modeling count data is Poisson distribution, quasi Poisson were carried out to test for over dispersion in the Poisson regression model and the formal way of dealing with over dispersion is negative binomial regression model, thus AIC was use to compare the two models, the Poisson regression model shows the best with minimum AIC. Furthermore optimal design were carried out using the optimality criterion that is the A and D optimality criterion, using design efficiency to compare the two (2) designs the optimality criterion with the highest efficiency is the best, thus D optimality criterion shows the best design.


References


Agresti, A., 2002. Categorical Data Analysis Wiley- Interscience, New York, USA.

Ash, A., Hedayat, A., 1978. An Introduction to Design Optimality with an Overview Literature. Communicat. Statist., 14, 1295-1325.

Atkinson, A.C., Chaloner, K., Herzberg, A.M., Juritz, J., 1993. Optimum Experimental Designs for Properties of a Comp. Mod. Biometr., 49, 325-337.

Atkinson, A.C., Donev, A.N., 1992. Optimum Experimental Designs, Clarendon Press, Oxford.

Atkinson, A.C., Donev, A.N., Tobias, R.D., 2007. Optimum Experimental Designs, with SAS. Oxford Univ. Press.

Box, G.E.P., Hunter, W.G., Hunter, S.J., 1978. Statistics for Experimenters, John Wiley & Sons, Inc., New York, NY.

Chernoff, H., 1953. Locally Optimal Designs for Estimating Parameters. Annal. Mathemat. Statist., 24, 586-602.

Fedorov, V.V., 1972. Theory of Optimal Experiments, Academic Press, New York.

Khuri, A.I., Mukherjee, B., Sinha, B.K., Ghosh, M., 2006. Design Issue for Generalized Linear Model. A Rev. Statist. Sci., 21,376-399.

Kiefer, J., 1959. Optimum Experimental Design (With Discussion). J. Roy. Statist. Soc. Ser. B., 21, 272-319.

McCullagh, P., Nelder, J.A., 1989. Generalized Linear Models 2nd Edition. London. Chapman Hall.

Montgomery, D.C., 2000. Design and Analysis of Experiments, Fifth Edition. John Wiley Sons, New York, NY.

Myers, R.H., Montgomery, D.C., Vining, G.G., 2002. Generalized Linear Models. John Wiley Sons, New York.

Nelder, J.A., Wedderburn, R.W.M., 1972. Generalized linear models. J. Roy. Statist. Soc., Series A 135, 370-384.

R Development Core Team., 2012. R: A language and environment for statistical computing. R Foundat. Statist. Comput., Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/.

Smith, K., 1918. on the standard deviation of adjusted and interpolated values of an observed polynomial Function and its Constants and the Guidance they give toward a proper cho. Distribut. Observat. Biometr., 12, 1-85

Wald, A., 1943. On the Efficient Design of Statistical Investigation. Mathemat. Statist., 14, 134-140.

Wheeler, R.E., 2004. optFederov. AlgDesign. The R project for statistical computing http://www.r-project.org/.

Yang, J., Mandal, A., Majumdar, D., 2012. Optimal designs for two-level factorial experiments with binary response. Statist. Sin., 22, 885–907.


Full Text: PDF

Refbacks

  • There are currently no refbacks.