In TaylorAndrew/atAnalyze: Methods for Iterative Analytical Reporting

Correlation Testing and Reporting:

corrCompare() computes a correlation table for each level of a binary grouping variable separately, and then tests differences in correlation coefficients between grouping variables

Usage: corrCompare(data = k, group = "grp", method = "pearson", rdec = 3, pdec = 3, tdec = 3)

corrStars() computes a correlation matrix, along with asterisks to indicate significance.

Usage: corrStars(x, method = "spearman", dec = 2)

Confidence Intervals:

multAovInteraction() is a wrapper for factor() which will apply a common label to one or more variables in data.frame.

Usage: ciMean(data = df, varlist = c("var1", "var2"), labels = c("No", "Yes"))

ciProp() reorders the ordering of the levels of a factor variable. If the order is not given, the factor levels are reverse-ordered.

Usage: ciProp(x, index_order = c(3, 1, 2))

Iterative Analysis and Reporting:

multAovInteraction() provides descriptive summary of multiple depedent variables, stratified by two grouping variables, along with p-values from aov() models testing the main effects and interaction between the two grouping variables.

Usage: multAovInteraction(data, grp_var1, grp_var2, varlist, round = 2, pround = 3))

multCox() will compute univariate Cox proportional hazards model models for multiple predictor variables.

Usage: multCox(data, timevar, event, vars, classvars = NULL, varLabelTable = NULL, dec = 2, pdec = 3)

multGroup() provides summary statistics for continous and categorical variables either overall or stratified by a grouping variable. It also provides parametric or nonparametric tests for each variable

Usage: multGroup(data, PcontinVars = NULL, PcatVars = NULL, NPcontinVars = NULL, NPcatVars = NULL, SortVars = NULL, varLabelTable = NULL, grouping = NULL, pdec = 3, dec = 2, mu = 0, ChiProbabilities = NULL, labels = NULL, percent = "column", NPdescriptives = NULL, padjust = F, provideP = T, include = "none", verbose = T)

multLogistic() performs univariate, multivariate, and/or conditional logistic regression models over a list of predictor variables.

Usage: multLogistic(data, y, predlist)

multOutcomesLogistic() provides results for univariate binary logistic regression models for a single predictor and multiple outcomes.

Usage: multOutcomesLogistic(data, predictor, outcomes)

multReg() performs univariate simple linear regression models for a single outcome variable over a list of predictor variables.

Usage: multReg(data, yvar, predlist)

t_testList() performs paired and unpaired parametric and non-parametric t.tests between columns within a data.frame, for multiple pairs of varaibles given in list form.

Usage: t_testList(data, list_of_t.tests

Other Survival Analysis Analysis Reporting Methods

kmTable() provides summary information from a kaplan-meier survival model, including median survival, event counts, and pairwise comparisons.

Usage: kmTable(data, time, event, group)

survPercent() provides survival percentages for a survfit(Surv()) model for specified times.

Usage: survPercent(model, times, labels = NULL)

Risk Difference Analysis

riskDifference() provides the risk difference for a categorical predictor and binary outcome. The reference/control group is the first level of the predictor variable.

Usage: riskDifference(data, y, x)

TaylorAndrew/atAnalyze documentation built on May 9, 2019, 4:21 p.m.