GLDEX: fun.bimodal.fit.pml – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

fun.bimodal.fit.pml

Finds the final fits using partition maximum likelihood estimation for the bimodal dataset.

Description

This is the secondary optimization procedure to evaluate the final bimodal distribution fits using the partition maximum likelihood. It usually relies on initial values found by fun.bimodal.init function.

Usage

fun.bimodal.fit.pml(data1, data2, first.fit, second.fit, prop, param1, param2, 
selc1, selc2)

Arguments

`data1`	First data set, usually obtained by `fun.class.regime.bi`.
`data2`	Second data set, usually obtained by `fun.class.regime.bi`.
`first.fit`	The distribution parameters or the initial values of the first distribution fit.
`second.fit`	The distribution parameters or the initial values of the second distribution fit.
`prop`	The proportion of the data set, usually obtained from `fun.bimodal.init`.
`param1`	Can be either `rs` or `fmkl`, depending on the type of first distribution used.
`param2`	Can be either `rs` or `fmkl`, depending on the type of second distribution used.
`selc1`	Selection of initial values for the first distribution, can be either `"rs"`, `"fmkl"` or `"star"`. Choose initial values from RPRS (ML), RMFMKL (ML) or STAR method.
`selc2`	Selection of initial values for the second distribution, can be either `"rs"`, `"fmkl"` or `"star"`. Choose initial values from RPRS (ML), RMFMKL (ML) or STAR method.

Details

This function should be used in tandem with fun.bimodal.init function.

Value

`par`	The first four numbers are the parameters of the first generalised lambda distribution, the second four numbers are the parameters of the second generalised lambda distribution and the last value is the proportion of the first generalised lambda distribution.
`value`	The objective value of negative likelihood obtained.
`counts`	A two-element integer vector giving the number of calls to functions. Gradient is not used in this case.
`convergence`	An integer code. `0` indicates successful convergence. Error codes are: `1` indicates that the iteration limit 'maxit' had been reached. `10` indicates degeneracy of the Nelder-Mead simplex.
`message`	A character string giving any additional information returned by the optimizer, or `NULL`.

Note

There is currently no guarantee of a global convergence.

Author(s)

Steve Su

References

Su (2007). Fitting Single and Mixture of Generalized Lambda Distributions to Data via Discretized and Maximum Likelihood Methods: GLDEX in R. Journal of Statistical Software: *21* 9.

Examples

## Extract faithful[,2] into faithful2
# faithful2<-faithful[,2]

## Uses clara clustering method
# clara.faithful2<-clara(faithful2,2)$clustering

## Save into two different objects
# qqqq1.faithful2.cc<-faithful2[clara.faithful2==1]
# qqqq2.faithful2.cc<-faithful2[clara.faithful2==2]

## Find the initial values
# result.faithful2.init<-fun.bimodal.init(data1=qqqq1.faithful2.cc,
# data2=qqqq2.faithful2.cc, rs.leap1=3,fmkl.leap1=3,rs.init1 = c(-1.5, 1.5), 
# fmkl.init1 = c(-0.25, 1.5), rs.leap2=3,fmkl.leap2=3,rs.init2 = c(-1.5, 1.5), 
# fmkl.init2 = c(-0.25, 1.5))

## Find the final fits
# result.faithful2.rsrs<-fun.bimodal.fit.pml(data1=qqqq1.faithful2.cc,
# data2=qqqq2.faithful2.cc, result.faithful2.init[[2]],
# result.faithful2.init[[3]], result.faithful2.init[[1]],param1="rs",
# param2="rs",selc1="rs",selc2="rs")

## Output
# result.faithful2.rsrs

GLDEX

Fitting Single and Mixture of Generalised Lambda Distributions (RS and FMKL) using Various Methods

v2.0.0.7

GPL (>= 3)

Authors

Steve Su, with contributions from: Diethelm Wuertz, Martin Maechler and Rmetrics core team members for low discrepancy algorithm, Juha Karvanen for L moments codes, Robert King for gld C codes and starship codes, Benjamin Dean for corrections and input in ks.gof code and R core team for histsu function.

Initial release

2020-02-04