SimDesign (version 2.18)

rtruncate: Generate a random set of values within a truncated range

Description

Function generates data given a supplied random number generating function that are constructed to fall within a particular range. Sampled values outside this range are discarded and re-sampled until the desired criteria has been met.

Usage

rtruncate(n, rfun, range, ..., redraws = 100L)

Value

either a numeric vector or matrix, where all values are within the desired range

Arguments

n

number of observations to generate. This should be the first argument passed to rfun

rfun

a function to generate random values. Function can return a numeric/integer vector or matrix, and additional arguments requred for this function are passed through the argument ...

range

a numeric vector of length two, where the first element indicates the lower bound and the second the upper bound. When values are generated outside these two bounds then data are redrawn until the bounded criteria is met. When the output of rfun is a matrix then this input can be specified as a matrix with two rows, where each the first row corresponds to the lower bound and the second row the upper bound for each generated column in the output

...

additional arguments to be passed to rfun

redraws

the maximum number of redraws to take before terminating the iterative sequence. This is in place as a safety in case the range is too small given the random number generator, causing too many consecutive rejections. Default is 100

Author

Phil Chalmers

Details

In simulations it is often useful to draw numbers from truncated distributions rather than across the full theoretical range. For instance, sampling parameters within the range [-4,4] from a normal distribution. The rtruncate function has been designed to accept any sampling function, where the first argument is the number of values to sample, and will draw values iteratively until the number of values within the specified bound are obtained. In situations where it is unlikely for the bounds to be located (e.g., sampling from a standard normal distribution where all values are within [-10,-6]) then the sampling scheme will throw an error if too many re-sampling executions are required (default will stop if more that 100 calls to rfun are required).

References

Chalmers, R. P., & Adkins, M. C. (2020). Writing Effective and Reliable Monte Carlo Simulations with the SimDesign Package. The Quantitative Methods for Psychology, 16(4), 248-280. tools:::Rd_expr_doi("10.20982/tqmp.16.4.p248")

Sigal, M. J., & Chalmers, R. P. (2016). Play it again: Teaching statistics with Monte Carlo simulation. Journal of Statistics Education, 24(3), 136-156. tools:::Rd_expr_doi("10.1080/10691898.2016.1246953")

See Also

runSimulation

Examples

# n = 1000 truncated normal vector between [-2,3]
vec <- rtruncate(1000, rnorm, c(-2,3))
summary(vec)

# truncated correlated multivariate normal between [-1,4]
mat <- rtruncate(1000, rmvnorm, c(-1,4),
   sigma = matrix(c(2,1,1,1),2))
summary(mat)

# truncated correlated multivariate normal between [-1,4] for the
#  first column and [0,3] for the second column
mat <- rtruncate(1000, rmvnorm, cbind(c(-1,4), c(0,3)),
   sigma = matrix(c(2,1,1,1),2))
summary(mat)

# truncated chi-square with df = 4 between [2,6]
vec <- rtruncate(1000, rchisq, c(2,6), df = 4)
summary(vec)