Motivation | Example | Comparisons | base package

Why a single hazard ratio may not be enough | Restricted mean survival time (RMST) | Reading and reporting RMST | Milestone (fixed-time) survival | Landmark analysis and immortal-time bias | Where to next

The inputs: a correlation matrix and its p-values | Basic correlogram | Reorder by clustering | Layout: triangles and mixed glyphs | Add the coefficients | Mark statistical significance | Colors and theme | It is a ggplot: compose freely | Computing p-values with cor_pmat() | Session information

Why this guide exists | Quick start: bring your own coordinates with as_factoextra_pca() | Non-linear embeddings (UMAP, t-SNE): use the dedicated faces | The two-layer architecture | The standardized return contract | Step 1 — Teach .get_facto_class() your class | Step 2 — Teach get_eig() your eigenvalues | Step 3 — Teach the get_*() extractor(s) | PCA | CA (and the mass helper you must not forget) | MCA | A complete worked example: ExPosition | A forward-looking template: vegan (RDA / CCA) | Testing your backend | Submitting a backend: checklist

Describe the data | Check the assumptions | Run the omnibus test | Post-hoc comparisons | Quantify the effect | Report it | Grouped analysis in one call | Plotting and further reading

Log-rank statistic for 2 groups | Weighted Log-rank extensions | Why are they useful? | Plots | Log-rank (survdiff) | Log-rank (weighted) | Gehan-Breslow (generalized Wilcoxon) | Tharone-Ware | Peto-Peto's modified survival estimate | modified Peto-Peto's (by Andersen) | Fleming-Harrington (p=1, q=1) | The weighted_logrank() function | References

Why not 1 − Kaplan-Meier? | Cumulative incidence, the right way | The naive-KM trap, made visible | Testing between groups: Gray's test | A number-at-risk table | Where to next

One model's coefficients | Is the effect the same in every subgroup? | One covariate across several models | Can you trust the hazard ratio? The proportional-hazards check | When the assumption fails | Where to next

Three lines to a survival curve | One argument to a publication-ready figure | Your arguments always win | Choosing a look | Reusable themes | The numbers every KM paper reports | Building on the curve | Saving your figure | Where to next

Dashed curves once few patients remain at risk | Step-shaped legend keys | Mark the median survival and its confidence interval | Plot math (expressions) in the legend | One shared legend across a grid of plots | Risk table under Cox-adjusted survival curves | Risk table under ggadjustedcurves() | Single event and a number-at-risk table for competing risks | Forest plot of several univariate Cox models | Smooth parametric (survreg / Weibull) survival curves | Subgroup-analysis forest (treatment effect within subgroups) | Compare parametric distributions against the Kaplan-Meier

Pairwise brackets | Automatic bracket packing | Compact letter display | Brackets or letters? | The omnibus label | All three in one call

From a table of estimates | From a fitted model | Mean differences and linear coefficients | Sorting, banding, and the estimate column | A meta-analysis forest

A single ROC curve | Which way round? The positive class and AUC below 0.5 | The optimal cut-point | Comparing several markers | Notes

The one-call two-way figure | It matches the by-hand analysis | Simple main effects on the plot | Customizing | Naming an effect on any plot

Side by side, stacked, and nested | Controlling the layout | One shared legend | Titles and panel tags | Correlation matrices | See also

The clustered correlogram | Lower triangle with coefficients | Marking significance | Circles for magnitude | Size-scaled squares | corrplot-style boxed cells | A colorblind-safe palette | A clean, edgeless heatmap | A rectangular predictor-by-outcome matrix | Going further: it is a ggplot | Session information

Introduction | Load required R packages | Compute a Cox model with interaction terms

Survival plot | Basic | Customized | Risk table | ncens plot | Notes