asbio (version 1.9-7)

rmvm: A multivariate normal dataset for data mining

Description

Contains a Y variable constrained to be a random function of fifteen X variables, which, in turn, are generated from a multivariate normal distribution with no correlation between dimensions.

Usage

data("rmvm")

Format

A data frame with 500 observations on the following 16 variables.

Y

A response vector defined to be: \(Y = X_1 + X_2 + X_3 + X_4 + X_5 + X_6 + X_7 + X_8 + X_9 + X_{10} + X_{11} + X_{12} + X_{13} + X_{14} + X_{15} + \epsilon\) where \(\epsilon \sim N(0, 1)\).

X1

A random predictor

X2

A random predictor

X3

A random predictor

X4

A random predictor

X5

A random predictor

X6

A random predictor

X7

A random predictor

X8

A random predictor

X9

A random predictor

X10

A random predictor

X11

A random predictor

X12

A random predictor

X13

A random predictor

X14

A random predictor

X15

A random predictor

Details

Data used by Derryberry et al. (in review) to consider high dimensional model selection applications.

References

Derryberry, D., Aho, K., Peterson, T., Edwards, J. (In review). Finding the "best" second order regression model in a polynomial number of steps. American Statistician.

Examples

## Code used to create data
if (FALSE) {
sigma <- matrix(nrow = 15, ncol = 15, 0)
diag(sigma) = 1
mvn <- rmvnorm(n=500, mean=rnorm(15), sigma=sigma)
Y <- mvn[,1] + mvn[,2] + mvn[,3] + mvn[,4] + mvn[,4] + mvn[,5] + mvn[,6] + mvn[,7] +
mvn[,8] + mvn[,9] + mvn[,10] + mvn[,11] + mvn[,12] + mvn[,13] + mvn[,14] + mvn[15] + rnorm(500)
rmvm <- data.frame(cbind(Y, mvn))
names(rmvm) <- c("Y", paste("X", 1:15, sep = ""))
}