B Testing

Description Usage Arguments Value Examples

Generates synthetic data from Gaussian distributions

1
2
3

SampleData(n.metrics = 100, n.observations = 20, mu.pre = 100,
  sigma.pre = 1, rho.ctrl = 0.8, rho.trmt = NULL, mu.ctrl = NULL,
  sigma.ctrl = NULL, mu.trmt = NULL, sigma.trmt = NULL, spread = FALSE)

`n.metrics`	Number of metrics.
`n.observations`	Number of observations.
`mu.pre`	Vector or scalar representing the mean during the pre-period for each metric.
`sigma.pre`	Vector or scalar representing the standard deviation during the pre-period for each metric.
`rho.ctrl`	Vector or scalar representing the correlation between the pre-period and the post-period for the control group for each metric.
`rho.trmt`	Vector or scalar representing the correlation between the pre-period and the post-period for the treatment group for each metric. If missing, set equal to rho.ctrl.
`mu.ctrl`	Vector or scalar representing the mean during the post-period for the control group for each metric. If missing, set equal to mu.pre.
`sigma.ctrl`	Vector or scalar representing the standard deviation during the post-period for the control group for each metric. If missing, set equal to sigma.pre.
`mu.trmt`	Vector or scalar representing the mean during the post-period for the treatment group for each metric. If missing, set equal to mu.pre.
`sigma.trmt`	Vector or scalar representing the standard deviation during the post-period for the treatment group for each metric. If missing, set equal to sigma.pre.
`spread`	If TRUE, the output has a horizontal structure with a `pre.post` column and column for each metric. If FALSE, the output has a vertical structure with a `metric` column, a `pre` column and a `post` column.

A data.frame which can be used as input for the PrePost and the PreCheck functions.

  data <- SampleData(n.metrics = 10)
  ans <- PrePost(data)

  n.metrics <- 10
  mu.pre <- 100
  mu.ctrl <- 100
  mu.trmt <- c(110, rep(100, n.metrics - 1))
  data <- SampleData(n.metrics = n.metrics,
                     mu.pre = mu.pre,
                     mu.ctrl = mu.ctrl,
                     mu.trmt = mu.trmt)
  ans <- PrePost(data)

  data <- SampleData(n.metrics = n.metrics,
                     mu.pre = mu.pre,
                     mu.ctrl = mu.ctrl,
                     mu.trmt = mu.trmt,
                     spread = TRUE)
  std.data <- ReshapeData(data)
  ans <- PrePost(std.data)