Expected Maximization Algorithm for the Estimation of   Missing Responses in Experimental Designs

Kuncham Srinivas; N. Ch. Bhatra Charyulu

doi:10.56093/jisas.v78i03.171400

Authors

Kuncham Srinivas University College of Science, Osmania University, Hyderabad
N. Ch. Bhatra Charyulu University College of Science, Osmania University, Hyderabad

https://doi.org/10.56093/jisas.v78i03.171400

Keywords:

RBD, LSD, RSD.

Abstract

This paper presents the estimation of missing responses in Randomized block Design (RBD), Latin Square Design (LSD) and Response Surface Design (RSD) using least squares method and expected maximization algorithm. The parameter relations also presented when initial values are taken as zero, mean of known responses and arbitrary. The methods are illustrated with suitable examples.

Downloads

Download data is not yet available.

References

Dempster, A.P., Laird, N.M., and Rubin, D.B. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B, 39, 1-38.

Draper, N.R. (1961). Missing values in Response Surface Designs. Technometrics, 3(3), 389-398.

Healy, M.J.R., and Westmacott, M. (1956). Missing values in experiments analyzed on automatic computers. Applied Statistics, 5, 203-206.

Nigam, and Gupta. (1979). Handbook on Analysis of Agricultural Experiments. IASRI Publications, New Delhi.

Srinivas Kuncham. (2018). Some contributions to estimation of missing values in design and analysis of experiments using expected maximization algorithm. Unpublished Ph.D. thesis, Osmania University.

Yates, F.Y. (1933). The analysis of replicated experiments when the field results are incomplete. Empire Journal of Experimental Agriculture, 1, 129-142.