Learn R Programming

⚠️There's a newer version (1.1) of this package.Take me there.

genpathmox (version 0.5)

Generalized Pathmox Approach Segmentation Tree Analysis

Description

Provides a very interesting solution for handling segmentation variables in complex statistical methodology. It contains en extended version of the "Pathmox" algorithm (Lamberti, Sanchez and Aluja,(2016)) in the context of Partial Least Squares Path Modeling including the F-block test (to detect the responsible latent endogenous equations of the difference), the F-coefficient (to detect the path coefficients responsible of the difference) and the "invariance" test (to realize a comparison between the sub-models' latent variables). Furthermore, the package contains a generalized version of the "Pathmox" algorithm to approach different methodologies: linear regression and least absolute regression models.

Copy Link

Version

Install

install.packages('genpathmox')

Monthly Downloads

241

Version

0.5

License

GPL-3

Maintainer

Giuseppe Lamberti

Last Published

November 16th, 2020

Functions in genpathmox (0.5)

Coefficient test

Defining labels of the binary partions.

Global test for the regression model

Defining the candidates to the optimum partition for each of segmentation variables (regression model)

Defining the matrices needed for the comparison test

Defining the matrices needed for the comparison test (regression model)

candidates.tree

Listing the candidates partions for each intermedied nodes

Defining the candidates to the optimum partition for each of segmentation variables

Coefficient test for the regression model

Check well defined blocks

Defining the linear relations between the latent variables.

Building the Bart matrix

get_path_scheme

Calculate inner weighting path scheme

Internal PLS regression (full data)

Create an indexed list from a dummy matrix

Listing the the f-coefficients test results for regression model for each obtained partition

Dummy by Giorgio

Listing the the f-coefficients test results for each obtained partition

Internal PLS regression with missing values

Check types of measurement scales and metric

Listing the the f-global test results for regression model for each obtained partition

Check bootstrap options

get_weights_nonmetric

Outer Weights Non-Metric Data

Calculate path coefficients for plspm

Check arguments for plspm and plspm.fit

Starting and ending positions

fglobal.tree.pls

Listing the the f-global test results for each obtained partition

Apply a function to all elements in a list

Check congruence between inner and outer models

Path coefficient effects for plspm

Get general parameters

Transform factors in MV into numeric

Calculate inner weighting path scheme

Split a data frame into blocks

Enumerating the possible combinations of a specified size from the elements of a vector

Check maximum number of iterations

Check specifications of PLS-PM algorithm

Check path matrix

Building data matrix with manifest variables

factor_to_dummy

Create a dummy matrix from the elements in a factor

fblock.tree.pls

Listing the the f-block test results for each obtained partition

Check vector of PLS components (for non-metric plspm)

Check tolerance threshold

Type of metric based on scaling measurement

Defining labels of a path coefficient.

Has missing values, NA, NaN, Inf

is_negative_decimal

Is negative decimal

Get data frame with bootstrap statistics

Has or lacks names?

is_square_matrix

Is square matrix

Defining vector minimum position

Performs bootstrap validation in plspm

Non-Metric Numerical Scale

Non-Metric Nominal Scale

Dummy matrices for categorical manifest variables

Product of all elements in a list

Goodness-of-fit for plspm

Has or lacks row/column names?

Create indices for elements in a vector or list

PATHMOX-REG: Segmentation Trees in linaer and LAD regression model

Non-Metric Dummy

get_inner_summary

Inner summary assessment

Same Number of Rows / Columns

Rank of a non-metric variable

Calculate Latent Variable Scores

Defining optimum partition for a specific variable (regression model)

get_treated_data

Apply corresponding treatment to MV

Is diagonal matrix

Unidimensionality of reflective blocks

Test if an object has one-dimension

Presence of missing values

invariance_test

Invariance Test

Test if an object is multi-dimensional

Calculating the invariance test

Sum of all elements in a list

is_negative_integer

Is negative integer

is_single_number

Is single number

Design-type matrix from the elements in a list

is_rectangular_matrix

Is rectangular matrix

list_with_vectors

List with vectors

is_single_negative

Is single negative number

Giving general information about the pathmox algorithm

is_single_logical

Is single logical

is_positive_decimal

Is positive decimal

is_positive_integer

Is positive integer

Giving general information about the pathmox-lm algorithm

create method treemodel

print.xtree.pls

Print function for the Pathmox Segmentation Trees: PLS-PM

is_single_decimal

Is single decimal

PLS-PM: Partial Least Squares Path Modeling

is_single_false

Is single false

is_single_positive_decimal

Is single positive decimal

is_single_positive

Is single positive number

is_lower_triangular

Is triangular matrix

is_square_numeric_matrix

Is square numeric matrix

test_manifest_scaling

Test scaling of selected manifest variables

Length of each element within a list

Resuming the tree attributes (regression model) : it gives general information about the tree

list_of_vectors

List of vectors

PATHMOX-PLS: Extended Segmentation Trees in Partial Least Squares Structutal Equation Modeling (PLS-SEM)

Normalize a vector

List with vectors of ones

Create a dummy matrix from the elements in a list

Calculating the element of all nodes

Create a list from a vector of integers

Resuming the tree attributes: it gives general information about the tree

test_null_weights

Test outer weights convergence within specified maxiter

Regression results of terminal nodes from the Pathmox Segmentation Trees

Defining the optimum partition given a set of segmentation variables

Defining binary partitions given a segmentation variable (factor).

Size: total number of elements in a list

PLS-PM results of terminal nodes from the Pathmox Segmentation Trees

test.partition.pls

Calculating the comparison tests

Calculating the element of the root node

Test presence of factors

is_single_negative_decimal

Is single negative decimal

create method xtree.pls

is_single_negative_integer

Is single negative integer

create method xtree.reg

print.xtree.reg

Print function for the Pathmox Segmentation Trees: linear regression and LAD

printing the tree structure

test.particion.reg

Calculating the comparison tests for the regression model

Calculating the element of the terminal nodes

is_single_positive_integer

Is single positive integer

create method treemodelreg

vector_to_dummy

Create a dummy matrix from the elements in a vector

is_single_string

Is single string

matrix_to_blocks

Split a matrix into blocks

Maximum of all elements in a list

Defining the optimum partition given a set of segmentation variables (regression model)

Mean of all elements in a list

Calculating size (numeber of individual of a node) stopo criterion

Minimum of all elements in a list

Calculating Deepth stop criterion

summary.xtree.pls

Summary function for the Pathmox Segmentation Trees: PLS-PM

summary.xtree.reg

Summary function for the Pathmox Segmentation Trees: linaer regression and LAD

Defining optimum partition for a specific variable.