splitMPP: Split MPP Data by Sliding Time Window
In mmpp: Various Similarity and Distance Metrics for Marked Point Processes

Description Usage Arguments Details Examples

This function splits a point process realization into a list of splitted point process realizations with length h.

1
2
3

splitMPP(mppdata, h = 60 * 60 * 48, ol = NULL, TimeOrigin = TRUE,
  scaleMarks = FALSE, scaleWindow = TRUE, MarkCenter = NULL,
  MarkCenterID = NULL)

`mppdata`	marked point process in data.frame composed of "time, mark1, mark2, ...".
`h`	width of the time window. Default is set to h=606048, which is two days when $time is recorded in second. This is suitable for a special seismic data only.
`ol`	length of overlap for the sliding window. Default 0.
`TimeOrigin`	logical. If `TRUE`, the beginning of the window is assumed to be the origin of time. Default `TRUE`.
`scaleMarks`	logical. If `TRUE`, marks (except time) are normalized to have unit variance in whole time series (not in individual windows). Default `FALSE`.
`scaleWindow`	logical. If `TRUE`, time interval (window.length) is normalized to one.
`MarkCenter`	vector for specifying the center of the mark. Use when there are relative center point such as the main shock of the earthquake. Default `NULL`.
`MarkCenterID`	vector for specifying the elements of center of the mark.

splitMPP splits a point process realization into a list of splitted point process realizations with length h.

##The aftershock data of 26th July 2003 earthquake of M6.2 at the northern Miyagi-Ken Japan.
data(Miyagi20030626)
## time longitude latitude depth magnitude
## split events by 5-hours
sMiyagi <- splitMPP(Miyagi20030626,h=60*60*5,scaleMarks=TRUE)