This function calls evaluate_solution, but since optim requires fn and gr to have the same parameters, it has two additional ones.
evaluate_solution.optim(par, data, evaluation_function = evaluate_solution, swap_treatment_function = NULL, mse_weights = NULL, change = NULL, prev_index_list = NULL)
a treatment assignment. The treatment and the data must have the same number of observations (rows).
a matrix containing the covariate vectors for each attribute.
the function used to evaluate the MSE treatment. Default is evaluate_solution, which does not take into account outcome or treatment weights. Other options are evaluate_solution_vector and evaluate_solution_matrix.
the parameter is only needed for optim, it does not play any role.
a vector containing the mse_weights for each treatment, or a matrix containing the mse_weights for treatments and outcomes and scaling factors.
Returns the mean square error value for the current treatment assignment.
Schneider and Schlather (2017),
ginv, optim
ginv
optim
# NOT RUN { input <- matrix(1:30, nrow = 10, ncol = 3) evaluate_solution.optim(par = c(0, 1, 1, 1, 1, 0, 0, 0, 0, 0), input) # }
Run the code above in your browser using DataLab