pcf_calc: Compute pair correlation function for breakpoint distribution
In jnghiem/bfasttools: Tools for BFAST spatial analysis

Description Usage Arguments Details Value Examples

This function computes the pair correlation function (PCF) for the point pattern of breakpoints. The PCF estimates g for a given radius distance r, where g represents the degree of clustering.

1 2	pcf_calc(bfast_in, level = 0.05, boundary, template, resample_dist = NULL, max.dist = NULL, output_directory = NULL, mc.cores = 1)

`bfast_in`	A data frame generated by `bfastSpatial()`.
`level`	A number indicating the significance level for the pointwise Monte Carlo test. See the documentation for `spatstat::envelope()` for more details. If `level=NA`, no envelope will be simulated.
`boundary`	A SpatialPolygons object defining the boundary of the PCF analysis. If this argument is not supplied, all records in `bfast_in` will be used.
`template`	A Raster object with the same resolution, extent, and projection as the raster dataset used to generate `bfast_in`. Alternatively, an XML file generated by `create_raster_metadata()`.
`resample_dist`	An optional number indicating the square dimension to which the breakpoint pattern raster should be resampled. This argument is in the same units as the projection of `template`. If not supplied, no resampling will occur and points will be generated based on the resolution of `template`.
`max.dist`	The maximum radius distance to which the PCF calculation should be performed. If not supplied, the default is the diagonal distance across `boundary`.
`output_directory`	An optional character file path to a directory to which the PCF calculations for each year and month should be written in `.rds` format. The radius distances for which the PCF was computed will also be written here.
`mc.cores`	A numeric indicating the number of cores to be used in parallel computing.

This function is a convenient wrapper for spatstat::pcf() and spatstat::envelope(). Most default values in pcf and envelope are accepted. However, the translation correction is used here.

If level is not set to NA, a simulation envelope will be computed according to the number of simulations needed to achieve level. However, the simulation process is very time-consuming so setting level=NA may be best choice for reasonable processing times.

The returned data frame has the following fields:

year: The year of the observation.
month: The month of the observation.
type: The type of statistic for g.
g: The value of g for the given type.
dist: The radius distance.
upper.bound, lower.bound: The upper/lower bound of the simulation envelope at distance dist.
nsim: The number of simulations to produce the envelope.
sig.level: The significance level of the pointwise Monte Carlo test.

A list of length 2. The first item is a data frame summarizing the PCF results. The second item is the vector of radius distances.

## Not run: 
pcf_calc(bf_df, NA, rgdal::readOGR(dsn="C:/Desktop/shapefiles", layer="Mojave"), template, 2000, "C:/Desktop/PCF_files", 6)

## End(Not run)