Create multivariate (correlated) data - for general distributions

Create multivariate (correlated) data - for general distributions

genCorFlex(n, defs, rho = 0, tau = NULL, corstr = "cs", corMatrix = NULL)

Arguments

n

Number of observations

defs

Field definition table created by function `defData`. All definitions must be scalar. Definition specifies distribution, mean, and variance, with all caveats for each of the distributions. (See defData).

rho

Correlation coefficient, -1 <= rho <= 1. Use if corMatrix is not provided.

tau

Correlation based on Kendall's tau. If tau is specified, then it is used as the correlation even if rho is specified. If tau is NULL, then the specified value of rho is used, or rho defaults to 0.

corstr

Correlation structure of the variance-covariance matrix defined by sigma and rho. Options include "cs" for a compound symmetry structure and "ar1" for an autoregressive structure. Defaults to "cs".

corMatrix

Correlation matrix can be entered directly. It must be symmetrical and positive semi-definite. It is not a required field; if a matrix is not provided, then a structure and correlation coefficient rho must be specified. This is only used if tau is not specified.

Value

data.table with added column(s) of correlated data

Examples

if (FALSE) {
def <- defData(varname = "xNorm", formula = 0, variance = 4, dist = "normal")
def <- defData(def, varname = "xGamma1", formula = 15, variance = 2, dist = "gamma")
def <- defData(def, varname = "xBin", formula = 0.5, dist = "binary")
def <- defData(def, varname = "xUnif1", formula = "0;10", dist = "uniform")
def <- defData(def, varname = "xPois", formula = 15, dist = "poisson")
def <- defData(def, varname = "xUnif2", formula = "23;28", dist = "uniform")
def <- defData(def, varname = "xUnif3", formula = "100;150", dist = "uniform")
def <- defData(def, varname = "xGamma2", formula = 150, variance = 0.003, dist = "gamma")
def <- defData(def, varname = "xNegBin", formula = 5, variance = .8, dist = "negBinomial")

dt <- genCorFlex(1000, def, tau = 0.3, corstr = "cs")

cor(dt[, -"id"])
cor(dt[, -"id"], method = "kendall")
var(dt[, -"id"])
apply(dt[, -"id"], 2, mean)
}