RMCP: R Model Context Protocol Server Documentation PyPI Version Downloads License: MIT Overview RMCP is a comprehensive Model Context Protocol (MCP) server implementing 40 statistical analysis tools across 9 categories. It enables AI assistants and applications to perform sophisticated statistical modeling, econometrics, machine learning, time series analysis, and data science tasks through natural language conversation. Quick Start Install RMCP Python package: Install required R packages (see R package requirements). Verify R packages: Start the MCP server: Features Statistical Analysis Tools (40 Tools) Regression & Correlation Linear regression (OLS with robust SE, R², p-values) Logistic regression (binary classification, odds ratios) Correlation analysis (Pearson, Spearman, Kendall) Time Series Analysis ARIMA modeling with forecasting Time series decomposition (trend, seasonal, remainder) Stationarity testing (ADF, KPSS, PP) Data Transformation Lag/lead variables Winsorization for outliers Differencing and standardization Statistical Testing T-tests (one-sample, two-sample, paired) ANOVA (Types I/II/III) Chi-square tests (independence/goodness-of-fit) Normality tests (Shapiro-Wilk, Jarque-Bera, Anderson-Darling) Descriptive Statistics Summary stats with grouping Outlier detection (IQR, Z-score methods) Frequency tables Advanced Econometrics Panel regression (fixed/random effects) Instrumental variables (2SLS, endogeneity tests) Vector autoregression (VAR) Machine Learning K-Means clustering (unsupervised clustering) Decision trees (classification/regression) Random forest (ensembles, variable importance) Data Visualization Scatter plots with trends Histograms with density overlay Box plots for quartiles/outliers Time series plots and regression diagnostics Correlation heatmaps File Operations CSV, Excel, JSON import/export Data filtering and info Natural Language & User Experience Formula builder from natural language Formula validation Intelligent error recovery and suggestions Data validation Example datasets for learning/testing Production Ready Full MCP protocol JSON-RPC 2.0 compliance Transport layer supports stdio, HTTP, WebSocket Safety: controlled R execution environment and secure error handling Real-World Usage Examples Business Analysis ROI analysis via regression: "Marketing spend yields $4.70 sales return per dollar with high significance (p < 0.001)" Economic Research Okun's Law validation via correlation: "GDP growth inversely correlates with unemployment, r = -0.944" Data Science Customer churn prediction with logistic regression: "Longer tenure reduces churn risk by 11.3% per month; model accuracy 100%" Natural Language Formula Building Converts natural language input to R formulas for regression: Sample formula: sales ~ marketingspend + customersatisfaction Intelligent Error Recovery Diagnoses missing R package errors and suggests fixes: Example: missing 'forecast' package, suggesting install.packages("forecast") Visual Analytics Plots (e.g., correlation heatmaps, scatter plots) appear directly in chat with Claude Combines statistics and professional-quality ggplot2 visualizations in conversation Installation & Setup Prerequisites Python 3.9+ R 4.0+ with required packages R Package Requirements