Quality API

Reusable financial data quality checks. These complement fs.profiler by exposing checks directly.

fs.quality.detect_gaps(df, freq="1bd")

Finds missing business dates. For multi-symbol DataFrames, returns a mapping of symbol to missing dates.

fs.quality.flag_outliers(df, method="zscore", threshold=4.0, columns=None)

Appends boolean outlier columns such as close_outlier.

Supported methods:

• "zscore"
• "iqr"

fs.quality.detect_splits(df, threshold=0.3, column="close")

Appends suspected_split, a boolean flag for large price jumps.

fs.quality.completeness_report(df)

Returns a JSON-serializable summary with row/column counts, gaps, null counts, duplicate rows, zero-volume days, outlier counts, and suspected split counts.