rho_M1Y: Approximate Correlation between Continuous Mixture Variable...

In SimCorrMix: Simulation of Correlated Data with Multiple Variable Types Including Continuous and Count Mixture Distributions

Description

This function approximates the expected correlation between a continuous mixture variables M1 and another random variable Y based on the mixing proportions, component means, and component standard deviations of M1 and correlations between components of M1 and Y. The equations can be found in the Expected Cumulants and Correlations for Continuous Mixture Variables vignette. This function can be used to see what combination of correlations between components of M1 and Y gives a desired correlation between M1 and Y.

Usage

rho_M1Y(mix_pis = NULL, mix_mus = NULL, mix_sigmas = NULL, p_M1Y = NULL)

Arguments

mix_pis	a vector of mixing probabilities that sum to 1 for component distributions of M1
mix_mus	a vector of means for component distributions of M1
mix_sigmas	a vector of standard deviations for component distributions of M1
p_M1Y	a vector of correlations between the components of M1 and Y; i.e., p_M1Y[1] is the correlation between the 1st component of M1 and Y

Value

the expected correlation between M1 and Y

References

Please see references for rho_M1M2.

See Also

rho_M1Y

Examples

# M1 is mixture of N(-2, 1) and N(2, 1); pairwise correlation set to 0.35
rho_M1Y(mix_pis = c(0.4, 0.6), mix_mus = c(-2, 2), mix_sigmas = c(1, 1),
  p_M1Y = c(0.35, 0.35))

SimCorrMix documentation built on May 2, 2019, 1:24 p.m.