Construction of Resolvable Spatial Row–Column Designs

Summary Resolvable row–column designs are widely used in field trials to control variation and improve the precision of treatment comparisons. Further gains can often be made by using a spatial model or a combination of spatial and incomplete blocking components. Martin, Eccleston, and Gleeson (1993, Journal of Statistical Planning and Inference34, 433–450) presented some general principles for the construction of robust spatial block designs which were addressed by spatial designs based on the linear variance (LV) model. In this article we define the two‐dimensional form of the LV model and investigate extensions of the Martin et al. principles for the construction of resolvable spatial row–column designs. The computer construction of efficient spatial designs is discussed and some comparisons made with designs constructed assuming an autoregressive variance structure.

[1]  R. J. Martin.  A subclass of lattice processes applied to a problem in planar sampling , 1979 .

[2]  H. D. Patterson,et al.  The efficiency of incomplete block designs in National List and Recommended List cereal variety trials , 1983, The Journal of Agricultural Science.

[3]  O. Mayo,et al.  Nearest Neighbour (Nn) Analysis of Field Experiments , 1983 .

[4]  Emlyn Williams,et al.  A criterion for the construction of optimal neighbour designs , 1985 .

[5]  Julian Besag,et al.  Statistical Analysis of Field Experiments Using Neighbouring Plots , 1986 .

[6]  Emlyn Williams,et al.  A neighbour model for field experiments , 1986 .

[7]  Brian R. Cullis,et al.  Residual maximum likelihood (REML) estimation of a neighbour model for field experiments , 1987 .

[8]  P. Wild,et al.  The construction of neighbour designs , 1987 .

[9]  Terence P. Speed,et al.  On a matrix identity associated with generalized least squares , 1990 .

[10]  EFFICIENT BLOCK DESIGNS FOR CORRELATED OBSERVATIONS , 1991 .

[11]  The empirical efficiency and validity of two neighbour models , 1991 .

[12]  Brian R. Cullis,et al.  Spatial analysis of field experiments : an extension to two dimensions , 1991 .

[13]  R. J. Martin,et al.  Robust linear block designs for a suspected LV model , 1993 .

[14]  R. Kempton,et al.  Statistical analysis of two-dimensional variation in variety yield trials , 1994, The Journal of Agricultural Science.

[15]  J. John,et al.  Cyclic and Computer Generated Designs , 1995 .

[16]  Sue J. Welham,et al.  Likelihood Ratio Tests for Fixed Model Terms using Residual Maximum Likelihood , 1997 .

[17]  J. Eccleston,et al.  Design Algorithms for Correlated Data , 1998, COMPSTAT.

[18]  J. John,et al.  t‐Latinized Designs , 1998 .

[19]  J. John,et al.  Recursive formulae for the average efficiency factor in block and row‐column designs , 2000 .

[20]  Gw Dutkowski,et al.  Analysis of early tree height in forest genetic trials is enhanced by including a spatially correlated residual , 2001 .

[21]  Roger W. Payne,et al.  Comparison of Nested Simulated Annealing and Reactive Tabu Search for Efficient Experimental Designs with Correlated Data , 2002, COMPSTAT.

[22]  Er Williams,et al.  Experimental Design and Analysis for Tree Improvement , 2002 .