# hedonic model: estimating coefficients for variables not used in the regression

I'm trying to estimate the value of a real estate upon its characteristics. To do so, I'm using the Hedonic Model and I'm doing the regression using lmRob (R package: robust).

There is a problem with the model and I don't know how to solve it.

If the real estate for which I want to evaluate the price, has more characteristics than the real estates used for the model creation, the evaluation wouldn't be complete.

Example:

Let's say that I want to evaluate a house which has 4 rooms, 190 m², an inside pool, 3 balconies and a sauna.

None of the other house from the neighborhood (houses which will be used for the regression) have an inside pool, the rest of the characteristics are similar.

This means that the house with the inside pool must have a bigger price than the other, but the regression will not take into account that the house has a inside pool.

My question is: how can I find the coefficient of the inside pool, in order to use it for the evaluation.

Thank you!

• Unfortunately it is quite difficult to estimate parameter values when you do not have data... Perhaps you can gather more from nearby neighbourhoods? – Analyst Mar 12 '15 at 12:18
• @Analyst Sadly I can't gather anything form the nearby neighborhoods. – Paul Mar 13 '15 at 12:35