Exploratory data analysis
in your browser

No install. No sign-up. No data upload.
Drop a CSV and start exploring.

MIDAS application showing a multi-pane workspace with data table, scatter plot, and statistical analysis

Graph Builder

Build any chart, layer by layer

Combine geometries and statistical transforms to build the chart you need — regression fits, prediction intervals, distributions, survival curves, faceted layouts, and more.

Scatter plot with regression lines and confidence bands

Time series with linear trend and prediction interval band

Kaplan-Meier survival curves with confidence intervals

Tile-based heatmap with sequential color scale

Scatter plot with regression lines and confidence bands

Statistical Models

Point-and-click modeling with built-in diagnostics

Configure and run models through a GUI. Results report confidence intervals and effect estimates rather than p-values. Regression models include residual diagnostics.

Ordinary Least Squares (OLS), Generalized Linear Model (GLM), Generalized Linear Mixed Model (GLMM), ANOVA, Principal Component Analysis (PCA), Random Forest, ARIMA, Cox Proportional Hazards, Kaplan-Meier, Design of Experiments (DoE), cross-tabulation, descriptive statistics.

GLM tab showing the configuration form with variable selection and the resulting coefficient table with confidence intervals

Residual diagnostic plots: residuals vs fitted, normal Q-Q, scale-location, residuals vs leverage

PCA results: scree plot showing variance explained and scatter plot of principal component scores

DoE main effects plot showing how each factor affects the response variable

ARIMA residual diagnostics: time series of residuals, Q-Q plot, ACF, and PACF correlograms

GLM — Configure a model and view coefficient estimates

SQL Editor

Shape your data with SQL

Write DuckDB SQL to filter, join, and transform datasets. Derived tables feed directly into models and graphs.

SQL editor with syntax-highlighted query and result table

Project Lineage

See how everything connects

Datasets, derived tables, models, and reports form a dependency graph. Project Lineage visualizes it so you can trace any result back to its source.

Dataset Derived Model Report

Project Lineage DAG showing datasets, derived tables, models, and reports with dependency edges

Your data never leaves your browser

MIDAS runs DuckDB compiled to WebAssembly inside your browser. All computation happens on your machine — no server receives your data. Install it as a PWA and work offline.

100% client-side No server processing Offline capable

Start exploring your data

Open MIDAS and drop your CSV file. No account required.

Launch MIDAS

Exploratory data analysisin your browser