> For the complete user guide index, see [llms.txt](https://portraitresearch.com/documentation/llms.txt).

# Standard quantitative filters

[Canonical page](https://portraitresearch.com/documentation/screening/standard-quantitative-filters/)

## Applicability

- **Task:** Understand quantitative filter categories and review requirements
- **Access:** Portrait account; Normal usage and concurrency limits may apply; source entitlements can apply
- **Audience:** Portrait users discovering companies with explicit criteria
- **Surfaces:** web; ios
- **Availability:** Production
- **Content type:** explanation
- **Last reviewed:** 2026-09-22

Use quantitative filters to narrow your company universe by financial metrics. On the web, add a filter from the builder, use the available metric list to compose an **Expression**, and optionally add a **Name**. Validate the expression before saving; validation uses one company, so review data coverage in the run results as well.

On iOS, ask the [AI Assistant](https://portraitresearch.com/documentation/screening/build-with-ai-assistant/#build-or-edit-a-screen-on-ios) for the metric, period, and threshold you need, then review and confirm the suggested filter. Use the web app when you need the manual expression editor.


## Market data

Use market-data filters for supported market-capitalisation, enterprise-value, or price-change constraints. Market-capitalisation and enterprise-value thresholds use raw US dollars in expressions. A return-since-date variable uses percentage points: `10` means a 10% price increase. See [Filter expressions, units, and missing data](https://portraitresearch.com/documentation/screening/filter-reference/) before mixing measures.

## Fundamental data

Supported historical fundamentals and consensus estimates can be filtered by absolute value, percentage change, compound annual growth, or revision. Coverage and forward periods depend on the metric and entitlement.

## Ratios and margins

Use this category for supported leverage, ratio, and margin conditions. If a required metric is absent, [request a coverage review](https://portraitresearch.com/documentation/more/custom-prompt-build/) rather than substituting a different measure without documenting it.

## Valuation metrics

Use valuation filters for supported multiples and valuation measures. Verify the metric definition and period before comparing companies.

## Recurring screens

The web quantitative editor has **Expression** and **Name** fields; it has no **Relative** date-range selector. Fiscal and rolling metric periods such as `F1` or `LTM` describe the data period and can change as data advances.

For a price-return lookback, ask the AI Assistant for an explicit rolling duration and verify the dated expression in each run. Web list/result reruns and scheduled runs refresh the date only for a supported rolling filter with exactly one return-since variable. **Run screen** from the web or iOS builder uses the saved dates without advancing them. An ordinary fixed dated expression is not automatically advanced.

Manual editing can stop a filter's date from rolling. After an edit, do not rely on the filter's name: inspect the dated expression in the next run. [Run and schedule screens](https://portraitresearch.com/documentation/screening/run-and-schedule-screens/) explains which actions refresh dates.