Advertisement

Designs in the Presence of Trends

  • Erkki P. Liski
  • Nripes K. Mandal
  • Kirti R. Shah
  • Bikas K. Sinha
Part of the Lecture Notes in Statistics book series (LNS, volume 163)

Abstract

Model: Fixed effects model for block designs with a first degree trend term specific to every block Optimality criteria: Universal optimality (UO) Major tools: Semi balanced arrays, Kiefer’s proposition 1 Optimality results: Classes of UO designs both within restricted subclasses and within the unrestricted class of designs Thrust: Combinatorial arrangement of treatments within each block of a BIBD so that the trend-adjusted C-matrix has maximal trace and completely symmetric structure

In this Chapter we consider a model for a block design where in addition to the block and treatment effect parameters, there is a linear trend term specific to every block. We first give designs which are optimal within the class of binary designs or within the class of trend free designs. We also give designs which are optimal within the unrestricted class. In all cases, we obtain trend-adjusted C-matrices which are completely symmetric and have maximal trace so that the designs are universally optimal. Semi balanced arrays of Rao are found very useful in this connection.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Bhaumik, D. K. (1993). On optimal block designs in the presence of a linear trend. Sankhya B 55, 91–102.MathSciNetzbMATHGoogle Scholar
  2. Bradley, R. A. and Yeh, C. M. (1980). Trend-free block designs: Theory. The Annals of Statistics 8, 883–893.MathSciNetzbMATHCrossRefGoogle Scholar
  3. Chai, F-Shun and Majumdar, D. (2000). Optimal designs for nearestneighbor analysis. Journal of Statistical Planning and Inference 86, 265–275.MathSciNetzbMATHCrossRefGoogle Scholar
  4. Cheng, C. S. (1988). A note on the optimality of semibalanced arrays, in Optimal Design and Analysis of Experiments, eds. Y. Dodge, V. Federov, and H. Wynn. North-Holland, Amsterdam, 115–122.Google Scholar
  5. Jacroux, M., Majumdar, D., and Shah, K. R. (1995). Efficient block designs in the presence of trends, Statistica Sinica 5, 605–615.MathSciNetzbMATHGoogle Scholar
  6. Jacroux, M., Majumdar, D., and Shah, K. R. (1997). On the determination and construction of optimal block designs in the presence of linear trends, Journal of the American Statistical Association 92, 375–382.MathSciNetzbMATHCrossRefGoogle Scholar
  7. Majumdar, D. (1998). Design and analysis in the presence of trends. Read at the conference Experimental Design: Theory and Applications held at Mathematics Research Institute, Oberwolfach, Germany.Google Scholar
  8. Martin, R. J. and Eccleston, J. A. (1991). Optimal block designs for general dependence structures, Journal of Statistical Planning and Inference 28, 67–81.MathSciNetzbMATHCrossRefGoogle Scholar
  9. Morgan, J. P. and Chakravarti, I. M. (1988). Block designs for first and second order neighbor correlations, Annals of Statistics 16, 1206–1224.MathSciNetzbMATHCrossRefGoogle Scholar
  10. Rao, C. R. (1961). Combinatorial arrangements analogous to OA’s, Sankhya A 23, 283–286.zbMATHGoogle Scholar
  11. Rao, C. R. (1973). Some combinatorial problems of arrays and applications to experimental designs, in A survey of Combinatorial Theory, eds. J. N. Srivastava, F. Harary, C. R. Rao, G. C. Rota and S. S. Srikhande, North-Holland, Amsterdam, 349–359.Google Scholar

Copyright information

© Springer Science+Business Media New York 2002

Authors and Affiliations

  • Erkki P. Liski
    • 2
  • Nripes K. Mandal
    • 1
  • Kirti R. Shah
    • 4
  • Bikas K. Sinha
    • 3
  1. 1.Department of StatisticsCalcutta UniversityCalcuttaIndia
  2. 2.Department of Mathematics, Statistics, and PhilosophyUniversity of TampereTampereFinland
  3. 3.Stat-Math DivisonIndian Statistical InstituteCalcuttaIndia
  4. 4.Department of Statistics and Actuarial ScienceUniversity of WaterlooWaterlooCanada

Personalised recommendations