We considered the adjustments of a set of nonlinear regression equations and the likelihood ratio test with approximations given by the χ2 and also by the F statistics, in order to test model identity and equality of any subset of parameters, using data with repetitions from a randomized complete block design. We concluded that both approximations can be used but the F should by preferred for small sample sizes.
Hypothesis test,; likelihood ratio; nonlinear equation comparisons
