Data Analysis ToolPak

This add-in provides data analysis and statistics functions, including probability distributions, hypothesis tests, summary statistics, regression models, and time-series analysis. These functions leverage widely used Python libraries such as SciPy and statsmodels. Since the functions are Python, they can be run using either Excel's integrated =PY() function or simply as a custom function which uses Pyodide locally in your browser. Either way you have full access to the Python source code for review and customization.

Features

🆓 Unlimited FREE use

🌐 Works in Excel for web and desktop

✅ Use Excel =PY() or local Python runtime

🔒 No data is shared outside Excel

🔍 Python source code available for review

📊 Statistical analysis functions

Functions

The following statistics function areas are included:

Probability Distributions

• Continuous, discrete, and multivariate distributions for density, cumulative probability, inverse CDF, survival functions, moments, and random variables.
• Common distributions such as normal, t, chi-square, F, beta, binomial, Poisson, gamma, lognormal, multinomial, Dirichlet, Wishart, and more.

Summary And Frequency Statistics

• Descriptive statistics, geometric and harmonic means, moments, skewness, kurtosis, modes, percentiles, and reliability metrics.
• Frequency tables, cumulative and relative frequency helpers, binned statistics, and percentile score calculations.

Hypothesis Tests

• One-sample, paired-sample, independent-sample, association, correlation, normality, variance, and goodness-of-fit tests.
• ANOVA, repeated-measures ANOVA, Welch ANOVA, post-hoc comparisons, and tools for comparing groups and validating assumptions.

Regression And Statistical Models

• OLS, WLS, GLS, robust regression, quantile regression, GLM, count models, discrete choice models, mixed effects, and survival models.
• Model diagnostics and helpers for estimating relationships between variables directly from worksheet ranges.

Bayesian Analysis

• Conjugate-prior updates, credible intervals, posterior summaries, tail probabilities, and weighted posterior summaries.
• Dirichlet-multinomial helpers for category probabilities, posterior updates, and predictive calculations.

Multivariate Analysis

• MANOVA, canonical correlation, factor analysis, PCA, multidimensional scaling, and random correlation matrices.
• Multivariate distribution helpers for workflows involving vectors, covariance matrices, and grouped measurements.

Time Series

• Autocorrelation, stationarity tests, moving averages, decomposition, and spectral analysis.
• Exponential smoothing, ARIMA/SARIMAX forecasting, order selection, and diagnostics for time-indexed data.

Use Cases

• Business Analytics: Summarize performance metrics, compare cohorts, test experiment results, and evaluate confidence intervals.
• Finance: Analyze returns, estimate distributions, run risk calculations, and model time-series behavior.
• Research: Run hypothesis tests, compare samples, fit models, and produce reproducible statistical calculations in spreadsheets.
• Operations: Analyze process variation, inspect quality metrics, forecast demand, and evaluate operational experiments.

For more details, see the detailed documentation.