Federer (1998, Biometrics 54, 471-481) presents two analyses of field data in which high-order polynomials are fitted as random regressions to remove spatial variation. We challenge the justification of this approach and suggest some alternatives.
No takes yet. Share an insight, caveat, or question.
A. Gilmour (2000) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: