easy_edm
easy_edm.RdThis is an automated workflow for performing causal analysis on the supplied time series using empirical dynamical modelling (EDM) techniques. It is intended to hide all the common steps of an EDM analysis, and should work on most datasets.
Usage
easy_edm(
cause,
effect,
time = NULL,
data = NULL,
direction = "oneway",
verbosity = 1,
showProgressBar = NULL,
normalize = TRUE
)Arguments
- cause
The causal time series (as a string or a vector).
- effect
The effect time series (as a string or a vector).
- time
For non-regularly sampled time series, the sampling times must be supplied here (as a string or a vector).
- data
If a dataframe is supplied here, then the cause, effect & time arguments must be the column names of the relevant time series as strings.
- direction
A string specifying whether we are checking a one directional causal effect or whether to test the reverse direction at the same time (work in progress!).
- verbosity
The level of detail in the output.
- showProgressBar
Whether or not to print out a progress bar during the computations.
- normalize
Whether to normalize the inputs before starting EDM.
Details
It may be the case that your data requires a custom analysis, so this function can be used as a helpful starting point from which to create a specialised analysis using the `edm` function directly.
Warning: While the `edm` functionality is well-tested and ready for use, this `easy_edm` automated analysis is still a work-in-progress.
Examples
library(fastEDM)
num_obs <- 500
map <- logistic_map(num_obs)
df <- data.frame(list(x = map$x, y = map$y))
x_causes_y <- easy_edm("x", "y", data=df, verbosity=0)
#> ✔ Strong evidence of CCM causation from x to y found.
y_causes_x <- easy_edm("y", "x", data=df, verbosity=0)
#> ✔ Strong evidence of CCM causation from y to x found.