Real estate · Econometrics

CityDataLab

Hedonic factor models that separate genuine price growth from changes in what is being sold.

A CityDataLab hedonic decomposition — like-for-like returns on each property attribute, 1995–2025
A CityDataLab hedonic decomposition — like-for-like returns on each property attribute, 1995–2025

CityDataLab applies econometrics to the property market. Headline price indices are misleading because the mix of what sells changes over time — a market can look like it is rising simply because larger or better homes happened to trade. CityDataLab’s hedonic factor models strip that out.

By modelling price as a function of a property’s attributes, it isolates the true, like-for-like return on each characteristic — floor area, freehold versus leasehold, energy rating and location — across London, New York, Paris and more.

Genuine growth, not composition

The core idea is to separate price growth that is real from price growth that is just a change in composition. Hedonic regression does exactly this, and CityDataLab makes the results legible across cities and time.

  • Hedonic regression on large open-property datasets.
  • Like-for-like return on floor area, tenure, energy rating and location.
  • Cross-city coverage — London, New York, Paris and beyond.
  • A clear separation of true growth from changes in the mix of sales.

Interested in CityDataLab?

Tell us about your problem — we’ll tell you how we’d approach it.

Get in touch