cross_lagged {scdtb}

Cross-Lagged Correlation

Description

Computes cross-lagged correlations between two variables in a dataframe.

Usage

cross_lagged(
  .df,
  .x,
  .y,
  lag.max = 5,
  na.action = stats::na.fail,
  conf.level = 0.95,
  ...
)

Arguments

.df

A dataframe containing the variables for analysis.

.x

The name of the first variable (as a string) to be analyzed.

.y

The name of the second variable (as a string) to be analyzed.

lag.max

The maximum lag at which to calculate the cross-correlation or covariance. Defaults to 5.

na.action

A function to specify the action to be taken if NAs are found. Defaults to stats::na.fail.

conf.level

The confidence level for the confidence intervals. Defaults to 0.95.

...

Additional arguments to be passed to stats::ccf().

Details

This function calculates the cross-lagged correlation between two variables in a given dataframe up to a specified maximum lag. It returns an object containing the cross-correlation function, confidence intervals, and other related information. The function calls stats::ccf() internally.

Value

An object of class "cross_lagged" containing the cross-correlation function, upper and lower confidence intervals, the number of observations used, and the names of the variables analyzed.

Examples

# Creating a sample dataset
reversal_withdrawal <- data.frame(
  phase = c(rep("baseline1", 6), rep("treatment1", 5), rep("baseline2", 5), rep("treatment2", 5)),
  time = 1:21,
  extbehavs = c(15, 10, 14, 17, 13, 12, 2, 1, 1, 0, 0, 9, 9, 11, 15, 20, 1, 0, 4, 0, 1)
)

reversal_withdrawal$synth <- sapply(reversal_withdrawal$time, function(x) {
  stats::rpois(1, x)
})

reversal_withdrawal <- as.data.frame(reversal_withdrawal)

# Using the cross_lagged function
cl_result <- cross_lagged(reversal_withdrawal, .x = "time", .y = "synth")