The package provides a framework for analyzing data from multivariate point processes in time or one-dimensional space, aka marked point processes with discrete marks, based on a specification of the conditional intensity process.
The current development version can be installed from the source in this
repository using the devtools
package.
install.packages("devtools") ## if 'devtools' not installed
library(devtools)
install_github("nielsrhansen/ppstat/pkg", build_vignettes = TRUE)
The best way to get started using ppstat is by working through the tutorial vignette.
library("ppstat")
vignette("Tutorial", package = "ppstat")
To see examples of how univariate and multivariate point processes can be fitted, you can check out the Toy Models vignette.
vignette("toyModels", package = "ppstat")
One important thing is that ppstat only fit models to data that are contained in an object of class MarkedPointProcess. Such objects can be constructed via the markedPointProcess function. The following vignette gives an introduction to the data structures that ppstat relies on.
vignette("Introduction", package = "ppstat")
See also the additional vignette on data containers and data visualization.
vignette("dataContainers", package = "ppstat")