On Testing Statistics for Comparing Several Measuring Devices

The main object of this paper is to compare the efficiency of measuring instruments that are used to measure the same quantity of interest in a group of $n$ individuals. The model considered was introduced in Grubbs (1948, 1973) and the inference for the model parameters is based on maximum likelihood estimation. A simulation study is used for comparing different test statistics for testing equality of biases and variances in the measuring instruments. A data set is used to illustrate the approach considered.