Sunflower Intelligence
Business analytics

Explore your data

Bring a table: a quality check, descriptive statistics, correlations, and regression with its diagnostics.

The room checks a table’s quality before any analysis, then describes each column, looks for relationships and fits a regression with its diagnostics. Your file never leaves your browser.

Your data

Paste a table or upload a CSV. Column types, gaps and unusual values are flagged, and nothing is changed without your say.

Dataset

Data quality

— / 100

    Suggested clean-up

      What has changed

        Columns
        First rows

        Describe

        Centre, spread and shape for any numeric column, with a 95% confidence interval for its mean.

        —mean

        Distribution

        Correlate

        How strongly do two columns move together? Pick the columns, then choose a cell to see the pair up close.

        Method

        Pearson measures straight-line relationships; Spearman ranks the values first, so it catches any steady rise or fall and resists outliers.

        Empty cells

        Pairwise keeps every row that has both values; listwise keeps only rows complete in every chosen column.

        Columns
        Correlation matrix

        Blue is positive, orange negative, grey near zero. Use the arrow keys to move between cells and Enter to open one.

        Pair

        Correlation is not causation. Two columns can move together because one drives the other, because a third factor drives both, or by chance. Treat a strong correlation as a question to investigate, not an answer.

        Regress

        Ordinary least squares: how much does the outcome change with each predictor, holding the others fixed, and can the model be trusted?

        Predictors
        —of the variation explained (R²)

        Rows used
        —
        Model F-test
        —
        Typical error (RMSE)
        —
        Mean absolute error
        —
        Coefficients
          Diagnostics

          Predict

          Enter values for each predictor. The range shows where 95% of individual outcomes should fall.

          Sources

          1. Breusch, T. S. & Pagan, A. R. (1979). A simple test for heteroscedasticity and random coefficient variation. Econometrica 47(5); studentized form after Koenker, R. (1981), Journal of Econometrics 17(1).
          2. MacKinnon, J. G. & White, H. (1985). Some heteroskedasticity-consistent covariance matrix estimators with improved finite sample properties. Journal of Econometrics 29(3). (HC3 standard errors.)
          3. Statistics here are computed in your browser and checked against SciPy and statsmodels in the site’s automated tests.
          4. All example datasets are synthetic, generated for this page.