Data methodology
How the Cape Verde property index is calculated.
CVREI tracks source-linked public property listings across Cape Verde. It publishes inventory and asking-price signals, not completed sale prices, valuations or an official market index. This page explains the population, calculations, suppression rules and limits behind the live market view and dated monthly briefings.
Methodology updated: 2026-08-23
01 · Scope
What the index measures.
The index describes the public listing market visible through monitored agency, broker and property-portal sources. A tracked listing is an advertisement, not a completed transaction.
The published population contains source-linked Cape Verde records that have passed the public-feed rules. Coverage is partial: private sales, off-market properties, direct transactions and unmonitored sources are outside the dataset.
02 · Population
From source listing to public record.
- Only published Cape Verde records with a source URL, at least one image and a supported island enter the canonical public feed.
- Superseded records and test sources are excluded from that public feed.
- Prices are represented in euros when a supported normalization is available. A missing price remains missing; CVREI does not estimate it.
- The feed uses canonical published records, but imperfect source data and residual duplicates may remain.
03 · Calculations
Two surfaces, two disclosed samples.
A median is the middle value after the eligible observations are sorted. The €/m² sample additionally requires a positive advertised floor area. Counts and source totals describe the snapshot population; they are not estimates of the entire Cape Verde market.
| Surface | Price sample | Publication threshold | Date |
|---|---|---|---|
| Live market overview | Numeric EUR asking prices from €10,000 to €5,000,000 | Island median shown at n≥5 | Updates with the live public feed |
| Monthly market briefing | Published canonical-feed records carrying a numeric EUR asking price | Asking-price and €/m² medians each require their own n≥10 sample | Frozen snapshot date printed on the edition |
04 · Publication rules
Small samples and concentration.
Monthly reports suppress a median when its relevant sample contains fewer than 10 observations. The asking-price and €/m² thresholds are evaluated separately, so one may be published while the other is withheld.
An edition may also withhold an island median when it would mainly describe one development rather than the island. There is not yet a universal numeric concentration threshold; any editorial withholding must be identified in that edition's methodology note.
Property-type comparisons use separately frozen segment samples. A blended island median should not be treated as a like-for-like comparison between islands with different property mixes.
05 · Limits
What these data cannot say.
- They cannot show completed sale prices, negotiation discounts or transaction volume.
- They cannot value an individual property or replace legal, financial or valuation advice.
- A change in the median may reflect a change in the mix of advertised properties, not a like-for-like change in property values.
- Coverage and source quality vary over time. Listing errors, stale advertisements and residual duplicates can affect the figures.
- The historical series is still short and does not support long-run market-cycle claims.
06 · Citation
How to cite a monthly edition.
Cite the report title, Cape Verde Real Estate Index (CVREI), the edition's publication date, its printed snapshot date and the permanent report URL. Quote the sample size beside any median. Link to this methodology page when the calculation or limitations matter.
Each edition carries its own methodology version and disclosures. If an edition-specific note conflicts with this overview, use the edition-specific note for that frozen snapshot.