Create an artificial model object

Create a pseudo-model object with the response variable being simulated according to the fixed and random effects. Model coefficients are replaced by the expectations specified in the argument beta. Variance-covariance components of random effects are replaced by the values specified in argument VarCov. The standard deviation of random error is replaced by the argument sigma. Creating such a pseudo-model facilitates power calculations by leveraging the anova function in lmerTest and the Anova function in car.

Usage

fit.pseu.model(formula, data, beta, VarCov, sigma, ...)

Arguments

formula

an object of class formula

data

a data frame with the independent variables of the design as columns, e.g., treatment factors and block factors.

beta

a vector of the expectations of model coefficients.

VarCov

variance-covariance matrices. If there are multiple random effect groups, supply the variance-covariance matrix of each group as an element in a list.

sigma

standard deviation of error

...

other arguments passed to the anova function in lmerTest. The type of sum of squares, with Type III as the default, and the method for computing the denominator degrees of freedom, with Satterthwaite's method as the default, can be changed. For more details, see anova.lmerModLmerTest.

Value

a pseudo model object.