Package 'dynRB'

Title: Dynamic Range Boxes
Description: Improves the concept of multivariate range boxes, which is highly susceptible for outliers and does not consider the distribution of the data. The package uses dynamic range boxes to overcome these problems.
Authors: Manuela Schreyer <manuelalarissa.schreyer@sbg.ac.at>, Wolfgang Trutschnig <Wolfgang.Trutschnig@sbg.ac.at>, Robert R. Junker <Robert.Junker@sbg.ac.at>, Jonas Kuppler <Jonas.Kuppler@uni-duesseldorf.de>, Arne Bathke <Arne.Bathke@sbg.ac.at>, Judith H. Parkinson <judith.parkinson@sbg.ac.at>, Raoul Kutil <Raoul_Hugo.Kutil@stud.sbg.ac.at>
Maintainer: Marco Tschimpke <marco.tschimpke@plus.ac.at>
License: GPL-2
Version: 0.18
Built: 2025-03-26 07:19:58 UTC
Source: CRAN

Help Index

Dynamic Range Boxes

Description

The package DynRB improves the concept of multivariate range boxes, which is highly susceptible for outlines and does not consider the distribution of the data. The package uses dynamic range boxes to overcome these problems.

Details

Package: dynRB
Type: Package
Version: 0.16
Date: 2021-05-11

Author(s)

Manuela Schreyer manuelalarissa.schreyer@sbg.ac.at,
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Arne Bathke Arne.Bathke@sbg.ac.at,
Judith H. Parkinson judith.parkinson@sbg.ac.at,
Raoul Kutil Raoul_Hugo.Kutil@stud.sbg.ac.at

References

Junker RR, Kuppler J, Bathke AC, Schreyer ML, Trutschnig W (2016) Dynamic range boxes - A robust non-parametric approach to quantify size and overlap of n-dimensional hypervolumes. Methods in Ecology and Evolution doi: 10.1111/2041-210X.12611

Judith H. Parkinson, Raoul Kutil, Jonas Kuppler, Robert R. Junker, Wolfgang Trutschnig, Arne C. Bathke: A Fast and Robust Way to Estimate Overlap of Niches and Draw Inference, International Journal of Biostatistics (2018)

Examples

# example function dynRB_VPa
# for reliable results use steps = 201
data(finch2)
r<-dynRB_VPa(finch2, steps = 101) 
r$result

Pairwise overlaps for each dimension

Description

Function returns pairwise overlaps for each dimension nn. Number of dynamic range boxes (steps) can be adjusted. Default: steps = 201

Usage

dynRB_Pn(A = A, steps = 201, correlogram = FALSE, row_col = c(2, 2))

Arguments

A

Data frame, where the first column is a character vector containing the objects (e.g. species) and the other columns are numeric vectors (containing measurements).

steps

Number of range boxes. Default: steps = 201

correlogram

If TRUE, the correlogram for each species is shown. If FALSE, no correlogram is shown. Default: correlogram = FALSE

row_col

Number of rows and columns of the figures (correlogram for each species). Default: row_col = c(2, 2)

Value

Data frame containing the summarized overlaps for each pair of objects and dimension.

Author(s)

Manuela Schreyer manuelalarissa.schreyer@sbg.ac.at,
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Arne Bathke Arne.Bathke@sbg.ac.at

References

Junker RR, Kuppler J, Bathke AC, Schreyer ML, Trutschnig W (2016) Dynamic range boxes - A robust non-parametric approach to quantify size and overlap of n-dimensional hypervolumes. Methods in Ecology and Evolution doi: 10.1111/2041-210X.12611

Examples

# example function dynRB_Pn
# for reliable results use steps = 201
data(finch2)
r<-dynRB_Pn(finch2, steps = 101)

Relative Dynamic Range Box size per dimension and object

Description

Function returns Dynamic Range Box size of each dimension nn. Number of dynamic range boxes (steps) can be adjusted. Default: steps = 201

Usage

dynRB_Vn(A = A, steps = 201, correlogram = FALSE, row_col = c(2, 2))

Arguments

A

Data frame, where the first column is a character vector and the other columns are numeric vectors.

steps

Number of range boxes. Default: steps = 201

correlogram

If TRUE, the correlogram for each species is shown. If FALSE, no correlogram is shown. Default: correlogram = FALSE

row_col

Number of rows and columns of the figures (correlogram for each species). Default: row_col = c(2, 2)

Value

Data frame containing the summarized niche length for each object and dimension.

Author(s)

Manuela Schreyer manuelalarissa.schreyer@sbg.ac.at,
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Arne Bathke Arne.Bathke@sbg.ac.at

References

Junker RR, Kuppler J, Bathke AC, Schreyer ML, Trutschnig W (2016) Dynamic range boxes - A robust non-parametric approach to quantify size and overlap of n-dimensional hypervolumes. Methods in Ecology and Evolution doi: 10.1111/2041-210X.12611

Examples

# example function dynRB_Vn
# for reliable results use steps = 201
data(finch2)
r<-dynRB_Vn(finch2, steps = 101)

Size and pairwise overlap

Description

Function returns size and pairwise overlaps of niches or trait-spaces. Size or overlaps of dimensions can be aggregated by using either "product", "mean" or "geometric mean" as aggregation method. The results obtained by using the product are automatically printed. Number of dynamic range boxes (steps) can be adjusted. Default: steps = 201

Usage

dynRB_VPa(A = A, steps = 201, correlogram = FALSE, row_col = c(2, 2), 
          pca.corr = FALSE, var.thres = 0.9)

Arguments

A

Data frame, where the first column is a character vector and the other columns are numeric vectors.

steps

Number of range boxes. Default: steps = 201

correlogram

If TRUE, the correlogram for each species is shown. If FALSE, no correlogram is shown. Default: correlogram = FALSE

row_col

Number of rows and columns of the figures (correlogram for each species). Default: row_col = c(2, 2)

pca.corr

If TRUE, a principal components analysis is performed.

var.thres

Variance predicted by the PCA-axes, if pca.corr = TRUE.

Value

Data frame containing the summarized niche overlap (and volume) for each pair of objects aggregated by all three possible choices (i.e. product, mean, geometric mean).

Author(s)

Manuela Schreyer manuelalarissa.schreyer@sbg.ac.at,
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Arne Bathke Arne.Bathke@sbg.ac.at

References

Junker RR, Kuppler J, Bathke AC, Schreyer ML, Trutschnig W (2016) Dynamic range boxes - A robust non-parametric approach to quantify size and overlap of n-dimensional hypervolumes. Methods in Ecology and Evolution doi: 10.1111/2041-210X.12611

Examples

# example function dynRB_VPa
# for reliable results use steps = 201
data(finch2)
r<-dynRB_VPa(finch2, steps = 101, correlogram = TRUE, row_col = c(1,1))
r$result

Data set finch

Description

To demonstrate the application of the functions for real world data, we used existing data sets on niches and trait-spaces and quantified their sizes and overlaps. The data set finch is a data set on morphological measurements of Darwin finches. The data set comprises quantitative measurements of nine traits characterizing five species of finches, each trait was measured at least in 10 individuals per species.

Usage

data("finch")

Format

A data frame with 146 observations on the following 10 variables.

Species

a character vector of the Species Geospiza heliobates, Geospiza prosthemelas prosthemelas, Geospiza fuliginosa parvula, Geospiza fortis fortis and Geospiza fortis platyrhyncha

BodyL

a numeric vector

WingL

a numeric vector

TailL

a numeric vector

BeakW

a numeric vector

BeakH

a numeric vector

LBeakL

a numeric vector

UBeakL

a numeric vector

N.UBkL

a numeric vector

TarsusL

a numeric vector

Source

Snodgrass R and Heller E (1904) Papers from the Hopkins-Stanford Galapagos Expedition, 1898-99. XVI. Birds. Proceedings of the Washington Academy of Sciences 5: 231-372.

Examples

data(finch)
## quick overview
head(finch)

Subset of data set finch

Description

To demonstrate the application of the functions for real world data, we used existing data sets on niches and trait-spaces and quantified their sizes and overlaps. The data set finch2 is a data set on morphological measurements of three Darwin finches. The data set comprises quantitative measurements of nine traits characterizing two species of finches, each trait was measured at least in 10 individuals per species.

Usage

data("finch2")

Format

A data frame with 103 observations on the following 10 variables.

Species

a character vector of the Species Geospiza fuliginosa parvula and Geospiza fortis fortis

BodyL

a numeric vector

WingL

a numeric vector

TailL

a numeric vector

BeakW

a numeric vector

BeakH

a numeric vector

LBeakL

a numeric vector

UBeakL

a numeric vector

N.UBkL

a numeric vector

TarsusL

a numeric vector

Source

Snodgrass R and Heller E (1904) Papers from the Hopkins-Stanford Galapagos Expedition, 1898-99. XVI. Birds. Proceedings of the Washington Academy of Sciences 5: 231-372.

Examples

data(finch2)
## quick overview
head(finch2)

Overview function

Description

This functions can be used to show the graphics generated by the functions dynRB_Pn,dynRB_Vn and dynRB_VPa.

Usage

overview(r, row_col = c(3, 3))

Arguments

r

Output of the function dynRB_Pn,dynRB_Vn or dynRB_VPa.

row_col

Number of rows and columns of the figures. Default: row_col = c(3, 3)

Author(s)

Manuela Schreyer manuelalarissa.schreyer@sbg.ac.at,
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Arne Bathke Arne.Bathke@sbg.ac.at

Examples

# example for the function dynRB_Pn
# for reliable results use steps = 201
data(finch2)
r<-dynRB_Pn(finch2, steps = 101)
overview(r)

Overlaps for each dimension using ranks

Description

Function returns the asymmetric overlaps for each dimension, calculated by the method published by Parkinson et al. (2018) using ranks. Further two confidence intervals are returned for each estimate. The confidence level, as well as the repetitions for bootstrap can be adjusted.

Usage

ranks_OV(A = A, alpha = 0.05, reps4boot = 1000, digit = 3)

Arguments

A

Data frame, where the first column contains two objects (e.g. species) and the other columns are numeric vectors (containing measurments).

alpha

The confidence level. Default: alpha = 0.05

reps4boot

Number of repetitions for the bootstrap. . Default: reps4boot = 1000

digit

Number of digits after which the results are cut off. Default: digit = 3

Value

Data Frame containing the two asymmetric overlaps for each dimension together with their confidence intervals. The last row contains the d-dimensional asymmetric overlaps.

Author(s)

Judith H. Parkinson judith.parkinson@sbg.ac.at,
Raoul Kutil Raoul_Hugo.Kutil@stud.sbg.ac.at,
Jonas Kuppler Jonas.Kuppler@uni-duesseldorf.de,
Robert R. Junker Robert.Junker@sbg.ac.at (corresponding author),
Wolfgang Trutschnig Wolfgang.Trutschnig@sbg.ac.at,
Arne Bathke Arne.Bathke@sbg.ac.at

References

Judith H. Parkinson, Raoul Kutil, Jonas Kuppler, Robert R. Junker, Wolfgang Trutschnig, Arne C. Bathke: A Fast and Robust Way to Estimate Overlap of Niches and Draw Inference, International Journal of Biostatistics (2018)

Examples

# example function ranks_OV
data(finch2)
head(finch2)
ranks_OV(finch2[1:4], alpha = 0.05)