Methodology

Evidence first, dates preserved, calculations inspectable and missing data left missing.

Last updated 6 September 2026

Source hierarchy

Altys prioritizes primary public sources such as exchange filings, company disclosures, regulatory releases and official datasets. Secondary sources can help locate or cross-check evidence, but they do not silently replace the retained source record.

Point-in-time treatment

Research data keeps the period it describes separate from the date it became available. Historical screens and backtests should use only information that was knowable on the evaluation date.

Calculations and comparability

Derived metrics follow documented, deterministic formulas and applicability rules. Consolidated statements are preferred when available, with standalone data used as a fallback. Bases are not mixed inside one calculation window.

Missing data

Absence is not treated as zero. If a value is not applicable, not computable or not covered, it remains unavailable and should carry an explanation where the product surface supports one.

AI and human judgment

AI can help retrieve, organize and summarize evidence. It must not invent a missing fact. Users remain responsible for assumptions, exceptions and investment decisions.

Limitations

Public pages are informational snapshots and may lag a new disclosure. Company data, classifications and calculations can be revised when stronger source evidence or a corrected processing rule becomes available. See the corrections process.