Miscellaneous Functions from M. Kohl

Contains several functions for statistical data analysis; e.g. for sample size and power calculations, computation of confidence intervals and tests, and generation of similarity matrices.

Statistical Classification

Performance measures and scores for statistical classification such as accuracy, sensitivity, specificity, recall, similarity coefficients, AUC, GINI index, Brier score and many more. Calculation of optimal cut-offs and decision stumps (Iba and Langley (1991), ) for all implemented performance measures. Hosmer-Lemeshow goodness of fit tests (Lemeshow and Hosmer (1982), ; Hosmer et al (1997), ). Statistical and epidemiological risk measures such as relative risk, odds ratio, number needed to treat (Porta (2014), ).

Power Analysis and Sample Size Calculation

Power analysis and sample size calculation for Welch and Hsu (Hedderich and Sachs (2018), ISBN:978-3-662-56657-2) t-tests including Monte-Carlo simulations of empirical power and type-I-error. Power and sample size calculation for Wilcoxon rank sum and signed rank tests via Monte-Carlo simulations. Power and sample size required for the evaluation of a diagnostic test(-system) (Flahault et al. (2005), ; Dobbin and Simon (2007), ) as well as for a single proportion (Fleiss et al. (2003), ISBN:978-0-471-52629-2; Piegorsch (2004), ; Thulin (2014), ), comparing two negative binomial rates (Zhu and Lakkis (2014), ), ANCOVA (Shieh (2020), ), reference ranges (Jennen-Steinmetz and Wellek (2005), ), multiple primary endpoints (Sozu et al. (2015), ISBN:978-3-319-22005-5), and AUC (Hanley and McNeil (1982), ).

Estimators of Prediction Accuracy for Time-to-Event Data

Provides a variety of functions to estimate time-dependent true/false positive rates and AUC curves from a set of censored survival data.

Tools for Statistical Disclosure Control in Research Data Centers

Tools for researchers to explicitly show that their results comply to rules for statistical disclosure control imposed by research data centers. These tools help in checking descriptive statistics and models and in calculating extreme values that are not individual data. Also included is a simple function to create log files. The methods used here are described in the "Guidelines for the checking of output based on microdata research" by Bond, Brandt, and de Wolf (2015) < https://cros.ec.europa.eu/system/files/2024-02/Output-checking-guidelines.pdf>.

Permutational Group Sequential Test for Time-to-Event Data

Permutational group-sequential tests for time-to-event data based on the log-rank test statistic. Supports exact permutation test when the censoring distributions are equal in the treatment and the control group and approximate imputation-permutation methods when the censoring distributions are different.

High Dimensional Categorical Data Visualization

Easy visualization for datasets with more than two categorical variables and additional continuous variables. 'diceplot' is particularly useful for exploring complex categorical data in the context of pathway analysis across multiple conditions. For a detailed documentation please visit < https://dice-and-domino-plot.readthedocs.io/en/latest/>.

Dice Plot Visualization for 'ggplot2'

Provides 'ggplot2' extensions for creating dice-based visualizations where each dot position represents a specific categorical variable. The package includes geom_dice() for displaying presence/absence of categorical variables using traditional dice patterns. Each dice position (1-6) represents a different category, with dots shown only when that category is present. This allows intuitive visualization of up to 6 categorical variables simultaneously.

A Time Series Database for Official Statistics with R and PostgreSQL

Archive and manage times series data from official statistics. The 'timeseriesdb' package was designed to manage a large catalog of time series from official statistics which are typically published on a monthly, quarterly or yearly basis. Thus timeseriesdb is optimized to handle updates caused by data revision as well as elaborate, multi-lingual meta information.

Parametric Time-to-Event Analysis with Variable Incubation Phases

Fit parametric models for time-to-event data that show an initial 'incubation period', i.e., a variable delay phase where the hazard is zero. The delayed Weibull distribution serves as foundational data model. The specific method of 'MPSE' (maximum product of spacings estimation) and MLE-based methods are used for parameter estimation. Bootstrap confidence intervals for parameters and significance tests in a two group setting are provided.