Today’s value is mistaken for a history
A screen can look sensible now while failing across earlier regimes or depending on a metric that was restated later.
Institutional screening
A stock screen should start the research, not end it.
Altys combines calculated quantitative filters with document evidence, point-in-time history and firm-specific research rules. The result is a shortlist that can move directly into company work, modelling, portfolio context and continuous monitoring.
Natural-language screeningCalculated fieldsDocument evidenceHistorical integrity
The operating problem
More information does not create continuous attention.
A retail screener is excellent at filtering today’s ratios. An institutional workflow needs to test when the rule would have worked, inspect the evidence and carry the result into the rest of the research process.
Qualitative evidence is disconnected
The number passes, but guidance, cash quality, segment mix or governance evidence may contradict the apparent signal.
The shortlist becomes another spreadsheet
After discovery, analysts manually rebuild the company context, model and monitoring rules in separate tools.
The Altys workflow
Research, decision and monitoring on one evidence trail.
Altys keeps the rule explicit and connects every pass to inspectable evidence.
01
Describe the economic idea
Begin with the investment logic in plain language, then translate it into explicit fields, definitions and thresholds.
02
Run calculated filters
Use code-owned financial, valuation, quality, growth, risk and factor metrics rather than numbers invented by a language model.
03
Add document conditions
Search filings, concalls and guidance for evidence that cannot be represented by a simple numeric field.
04
Test through time
Evaluate the rule point-in-time, including sector and regime behaviour, turnover, drawdown and realistic availability.
05
Continue the workflow
Open a passing company in the research board, model the drivers, write the case and monitor the conditions that caused it to pass.
What the system adds
Infrastructure around the analyst, not a black box above them.
The language model should help formulate the question. Deterministic calculations and sourced documents should answer it.
Natural language, explicit logicAI helps express the idea, while the final quantitative rule remains inspectable.
Point-in-time testsHistorical membership reflects information available on each date.
Thesis-based screeningCombine financial thresholds with sourced qualitative evidence and firm-specific scorecards.
Live forward trackingKeep watching the screen after the backtest so the strategy can build an honest out-of-sample record.
Questions, answered
What teams usually ask before a pilot.
What makes a stock screener institutional?
An institutional screener needs consistent definitions, point-in-time testing, source lineage, portfolio and risk context, company-specific evidence and a path from discovery into the research and monitoring workflow.
Can Altys screen qualitative information?
Altys can search and structure evidence from filings, concalls and guidance alongside calculated quantitative fields. The underlying source remains available for review.
Does natural-language screening mean AI calculates the ratios?
No. AI can translate an investment idea into candidate conditions. Financial ratios, growth, valuation and risk fields should still be calculated deterministically from sourced data.
Can the same screen be monitored after launch?
Yes. A historical strategy can continue as a live forward test, and companies can trigger alerts as they enter, leave or approach the defined conditions.
Private preview
Research once. Let Altys keep watching what changes.
Bring a real company universe and a real monitoring problem. We will show how Altys can connect the data, research workflow and alerts around the way your team already invests.