processdata-package: Process Data

In ppstat: Point Process Statistics

Description

This package implements data structures for stochastic process data. The data can be in the form of either discretely observed continuous time processes, marked point processes, jump processes or a combination.

Details

Package:	processdata
Type:	Package
Version:	0.8.2
Date:	2013-04-02
License:	GPL version 2 or newer
LazyLoad:	yes

The class ContinuousProcess implements a data structure for observations from a potentially multivariate continuous time process observed at discrete time points, which are not necessarily equidistant. The class MarkedPointProcess extends this class to include observations of marked point processes. A further extension is the class JumpProcess, which uses the combination of marked points and continuous process observations to represent data from jump processes.

The classes are, in particular, designed as a convenient data format for storing date from multiple units or individuals where we have multivariate or even high-dimensional observations of processes, and where the processes can be a combination of discretely observed continuous processes as well as discrete events. The classes manage the relations between the different types of data and provide a simple interface but avoids memory expensive redundancies in the representation.

For an introduction to the usage of the package, see the Introduction vignette.

Author(s)

Niels Richard Hansen [e-mail].

Maintainer: [e-mail]

Examples

### Setting up a simple continuous time process with equidistant 
### sampled observations.

CPframe <- data.frame(time = seq(0,1,0.01),
                      val1 = rnorm(101,log(1:100)),
                      val2 = cumsum(rnorm(101,-0.1)))

CP <- continuousProcess(CPframe)

## Not run: 
plot(CP)
## End(Not run)

### Setting up a simple point process with two marks
PPframe <- data.frame(
 time = c(0.1,0.12,0.35,0.5,0.55,0.56,0.7,0.75,0.9,0.91),
 markType = c("A","A","B","B","A","A","B","A","B","A")
)

PP <- markedPointProcess(PPframe, seq(0,1,0.01))


## Not run: 
plot(PP)
## End(Not run)

### Combining the two data structures in a single object
process <- markedPointProcess(PPframe, CPframe)

## Not run: 
plot(process)
## End(Not run)


## More details in the vignette.
## Not run: 
vignette("Introduction", package = "processsdata")
## End(Not run)

ppstat documentation built on May 2, 2019, 5:26 p.m.