mcp: Multiple change-point analysis for layer boundary...

In maxgav13/GMisc: Miscellaneous Functions for Geology (and its areas)

Multiple change-point analysis for layer boundary determination

Description

Performs multiple change-point analysis in univariate or multivariate data, to find layer boundaries in a perforation log.

Usage

mcp(
  data,
  R = 199,
  alpha = 2,
  sig.level = 0.01,
  min.perc = 15,
  conf.level = 0.95
)

Arguments

data	A data frame containing the depth/distance in the first column, and the variables of interest in the rest of the columns, for a CPTu test: point resistance (qc), sleeve friction (fs), and pore-water pressure (u)
R	The number of random permutations
alpha	A parameter between 0 (exclusive) and 2 (inclusive). lower values make allow for more variation in the search for changepoints
sig.level	Significance level to determine significance of the changepoints
min.perc	Minimum percentage of data points per layer (between changepoints)
conf.level	Confidence level to use for plot and summary statistics (Default is 0.95)

Details

The example data given is intended to show the structure needed for input data. The user should follow this structure, which in general corresponds with a data frame with a sequence in the first column and the observed/measured values in the rest of the columns

Value

ggplot and plotly objects showing the layer distinction, statistical summary of the layers, and a summary table

References

Nicholas A. James, David S. Matteson (2014). ecp: An R Package for Nonparametric Multiple Change Point Analysis of Multivariate Data, Journal of Statistical Software, 62(7), 1-25.

Examples

mcp(CPTu_data, R = 199, alpha = 2, sig.level = .01, min.perc = 15) # multivariate example
mcp(DPM_data, R = 199, alpha = 2, sig.level = .01, min.perc = 15) # univariate example

maxgav13/GMisc documentation built on June 12, 2022, 3:48 a.m.