Robust analysis of agricultural field experiments

Abstract

Agricultural data generated from designed experiments are also prone to occurrence outliers. It is well known that Least Squares (LS) model can be distorted even by a single outlying observation. An outlier is one that appears to deviate markedly from the other members of the sample in which it occurs. The sources of influential subsets are diverse. Rousseeuw (1984) introduced a robust method known as Least Median of Squares (LMS) for linear regression models. By this method, the median of squares errors is minimized in order to obtain parameter estimates. It turns out that this estimator is very robust with respect to outliers. Since it focuses on the median residual, up to half of the observations can disagree without masking a model that fits the rest of the data. Therefore, the breakdown point of this estimator is 50%, the highest possible value. In the present investigation, this method is applied to analyze the data set containing outlying observations generated from agricultural field experiments. The data sets for the present investigation have been taken from Agricultural Field Experiments Information System, IASRI, New Delhi.