Rmad: Generate a random R matrix with given MAD from a...

In fungible: Psychometric Functions from the Waller Lab

Generate a random R matrix with given MAD from a user-defined, population R

Description

Rmad(Rpop, MAD = .10, tries = 1E5, SEED = NULL)

Usage

Rmad(Rpop, MAD = 0.03, tries = 1e+05, SEED = NULL)

Arguments

Rpop	(matrix) A model-implied, population correlation matrix.
MAD	(scalar) The desired mean absolute deviation value (i.e., mad(Rpop - R) = MAD where mad computes the mean absolute deviation for the nonredundant, non-diagonal values of its argument).
tries	(scalar) The maximum number of tries for the optimization algorithm. Default(tries = 1E5).
SEED	(numeric) The starting seed for the random number generator. If SEED is not supplied then the program will generate (and return) a randomly generated seed.

Value

• R A random R matrix with a known MAD from Rpop.

• Rpop The model-implied, population correlation matrix.

• Delta The vector of model-errors.

• H A hollow matrix used to create a fungible R matrix.

• iter Iteration number at convergence (or max iteration number).

• MAD The target mean absolute deviation.

• converged. A Boolean that describes the convergence status of the optimization algorithm.

• SEED The initial value for the random number generator.

Author(s)

Niels Waller

Examples

 # Example 1
  Rpop <- matrix(.35, 6, 6)
  diag(Rpop) <- 1
  
  out <- Rmad(Rpop, MAD = .038, SEED = 123)
               
  out$R |> round(3)            
 
# E = mattix of model error values
  E = Rpop - out$R 
  mean(abs(E[lower.tri(E, diag = FALSE)]))
   

fungible documentation built on April 7, 2026, 9:06 a.m.