A R package developed for the 6. International Summer School "Human Biology and Public Health - Data Analysis and Statistics" of the University of Potsdam in Gülpe in the North-East of Brandenburg.
The package contains a few functions required for reliable data analysis which are not part or only in more basic format in standard R.
The following methods are provided within the athi-environment from version 0.4.1:
athi$assoc_plot- assocplot with statistics values and Pearson residual shading (plot)athi$box_plot- boxplot with statistics values (plot)athi$bootstrap- resampling for the given data set and function (stats)athi$cdist- correlation distances (stats)athi$chr2ord- map characters or factors to numbers with comfort (data)athi$cihist- confidence interval for histogram density plots (plot)athi$cohensD- effect size for difference between two means (stats)athi$cohensW- effect size for categorical data (stats)athi$corr- pairwise correlations and their statistics (stats)athi$cor_plot- extended xy-plot with statistics (plot)athi$df2md- print a matrix or data frame as a Markdown table (document)athi$drop_na- drop rows if the given columns contain NAs in this row for certain columns (data)athi$epsilon_squared- effect size measure for Kruskal test (stats)athi$eta_squared- effect size measure for ANOVA (stats)athi$fmt- formatted string output (data)athi$impute- impute missing values using rpart or knn methods (data)athi$input- readline replacement for scripts (scripts)athi$introNAs- introduce missing values (data)athi$kroki- draw diagrams and flowcharts using kroki webservice (plot)athi$lm_plot- plot a linear model with confidence intervals (plotathi$mds_plot- plot a multidimensional scaling analysis (plot)athi$mi- determine mutual information (stats)athi$norm- normalize data (data)athi$pastel- create up to 20 pastel colors (plot)athi$pca_biplot- improved biplot for pca objects (plot)athi$pca_oncor- perform a PCA on a square (correlation) matrix (stats)athi$pca_pairs- improved pairs plot for pca objects (plot)athi$pca_plot- improved screeplot for pca objects (plot)athi$pcor- measuring partial correlation (stats)athi$pcor.test- test for significance of partial correlation (stats)athi$randomize- randomize column data within matrix or data frame (data)athi$ref_score- WHO reference score for the given age, sex and type (data)athi$ref_table- reference tables for WHO for the given sex and measure type (data)athi$report_pvalue- report p-values within documents using default alpha thresholds (document)athi$sem- standard error of the mean (stats)athi$smartbind- combine two data frame even if the have different column names (data)}athi$textplot- write the data for a data frame or matrix into a plot (plot)athi$untab- expand a contingency table to a data frame one item per row (data)athi$venn- Venn diagram for logical relations between two and three sets (plot)athi$wilcoxR- effect size r, for a wilcox test object (stats)
The package can be directly installed from the Github repository like this: Start R and install the package with the following command:
install.packages(
"https://github.com/mittelmark/athi/releases/download/0.4.0/athi_0.4.1.tar.gz",
repos=NULL)To install the latest development version directly from the Github repository
you need the R package remotes to be installed, if you have this package you can
do the following:
library(remotes)
remotes::install_github("https://github.com/mittelmark/athi")
You should do this if you need functions from the listing above which are currently only on Github, but not in the latest release.
The package comes with the usual documentation and vignettes.
To see the main entry page for the documentation use the library help function like this:
library(athi)
library(help="athi")
To browse the vignette(s) use the vignette command like this:
library(athi)
vignette("athi-functions",package="athi")
Author: Detlef Groth, University of Potsdam, Germany
License: MIT License see the file LICENSE for details.
In case of bugs and suggestions, use the issues link on top.