Анализ данных и регрессия на R

data

data-analysis is a Claude Code skill that runs an end-to-end R data analysis pipeline — from loading a dataset through exploratory analysis, cleaning, and regression to publication-ready tables and figures. The workflow spans five phases: a Pre-Flight Report that reads variable names, types, and missingness rates before writing a single line of analysis code; EDA with histograms, scatter plots, and group comparisons; regression using fixest for panel data or lm/glm for cross-sections with properly clustered standard errors; and output generation via modelsummary (.tex and .html tables) and ggplot2 (.pdf and .png figures). Scripts are saved to scripts/R/ and all outputs to output/; a built-in r-reviewer agent checks the generated script before results are presented. Long-running fits are launched in the background using the Monitor tool. Ideal for researchers and analysts who need reproducible empirical results from a CSV, RDS, or DTA file.