Sample Size Estimation of Nonparametric Tests with Ordered Alternatives for Longitudinal Data in Randomized Complete Block Designs

Longitudinal studies involve repeated measurements from the same subjects or blocks over short an extended periods of time. In longitudinal studies, usually most important step is to decide how many experimental units use. There are no closed form equations for determining sample size in complex designs. Monte Carlo simulation method effective tool designs estimate power size. This paper introduces estimating number based on a fixed treatment/time randomized complete block with correlated responses analyzed by nonparametric tests against ordered alternatives. The estimated each test statistics taking into account autocorrelation structure error terms which either stationary first-order moving average autoregressive non-normally distributed white noise terms. An extensive size/power comparison among recently proposed Modification S and other two well-known such as Page generalized Jonckheere alternatives carried out under structures Laplace Weibull distributions. Simulation study indicates that distribution have role estimation test. requires large contrast data specified setting.