Methodology of real estate and demographic data in Loiret

Each sale in the DVF database (Demandes de valeurs foncières, published by DGFiP) is converted into a price per square meter:

price per m² = land value ÷ actual built area

We then group sales by municipality, year, and property type (house or apartment), and then retain the median : the value that divides the series into two equal halves.

median = central value of the sorted series (50% of sales below)

The median is preferred over the mean because it resists extreme values: an exceptional property or a sale at a symbolic price does not pull the result up or down.

Excluded from the calculation are: sales without specified built area, bare land, commercial premises, multiple sales (a single deed covering several properties), and outlier prices per m² (extremely low or high values, typically outside the range of 100 – 15,000 €/m²).

The change compares two annual medians for the same territory and the same property type:

change (%) = (median N − median N−1) ÷ median N−1 × 100

Example: a municipality going from 1,500 €/m² to 1,620 €/m² displays +8.0%. The gap between two territories expressed as a percentage is presented in points (pts), so as not to confuse "percentage gap" and "percentage change".

A change is only displayed if the two compared years pass the reliability threshold. In small municipalities, a sharp change often reflects a shift in the nature of properties sold (a few renovated houses one year, older buildings the next) rather than a true market trend.

A median price is only published if the municipality has at least 5 sales of the same property type during the year under review. Below this threshold, the median becomes unstable: with two or three transactions, a single atypical sale shifts the result by several hundred euros per m².

In practical terms, on the site's pages: municipalities below the threshold display "—" instead of a price, they are excluded from rankings and departmental averages, and change rankings offer a reliability filter requiring 5 sales in each of the two compared years.

The number of sales is systematically displayed next to the price: it is the best indicator of the figure's reliability. A median calculated on 6 sales remains indicative; a median on 200 sales is robust.

At the scale of an intercommunality or the department, the median is recalculated across all sales in the territory, and not obtained by averaging municipal medians: this avoids giving as much weight to a municipality of 300 inhabitants as to Orléans.

In comparison tools, the "Average" line is, on the other hand, a simple and unweighted average of the territories you selected; "Max gap" is the difference between the highest and lowest value of the selection. The radar chart normalizes each indicator from 0 (lowest value in the selection) to 100 (highest): it is used to read relative positions, not absolute values.

Population data comes from INSEE legal population figures, whose vintage is offset by three years (the population 2024 is published later): population and prices displayed side by side do not always cover the exact same year.

• DVF does not tell everything about the property. Condition, renovations, floor, exposure, garden, garage, and energy performance are not in the database. Two municipalities with the same median price can offer very different properties.
• Composition effect. A year dominated by sales of new builds or renovated houses drives up the median without the "existing" market having moved.
• Partial coverage. VEFA sales (buying off-plan), certain transfers of real estate company shares, and properties with no specified area are missing or excluded. Alsace-Moselle and Mayotte are not covered by DVF, which does not affect Loiret but limits national comparisons.
• Time lag. Deeds are published semi-annually with a delay of several months: the last available year (2024) does not reflect the market of the very last few weeks.
• Small sample sizes. Even above the threshold of 5 sales, rural municipalities retain sensitive medians; compare at the intercommunality scale instead.
• This is not a property valuation. These figures describe a market, not the value of a specific home. For an appraisal, contact a professional.

All data used is reusable public data (open license). We republish calculations, not individual deeds. If you notice an inconsistency, please report it: processing is recalculated with each new DVF publication (2021 – 2024).

Municipalities & real estate prices in LoiretFactsheets for the 325 municipalities