R/dk_sim.R
dk_sim.RdBackward, forward and both automated subset selection algorithms: Frequency of obtaining authentic and noise variables by Shelley Derksen H. J. Keselman, British Journal of Mathematical and Statistical Psychology (1992) 45, 265-282.
dk_sim(file = "", directions, betas, n_values, alpha_values, rho_values, predictors, weight = 1, nlambda = 100, sims, error_fun, ...)
| file | character containing file argument to cat used to report intermediate results |
|---|---|
| directions | character vector with one or more of "backward", "forward", or "both". |
| betas | coefficients of regression |
| n_values | vector of sample sizes |
| alpha_values | vector of alpha threshold values |
| rho_values | vector of correlations between real predictors |
| predictors | vector of the number of predictor variables to include |
| weight | observation weights. Can be total counts if responses are proportion matrices. Default is 1 for each observation |
| nlambda | The number of lambda values - default is 100. |
| sims | number of simulations |
| error_fun | function providing the error term for each row of observations |
| ... | additional arguments need by |
object with
Need to allow add and remove thresholds instead of a single value.