Value of Features and Attributes to Statistically Estimate Rental Unit Pricing

#1
Let me start by saying I don't have much of a background in statistics, but I am attempting to inject some statistical analysis into a relatively straightforward pricing model that I use to price rental units. I'm just looking for a place to start, so any help would be appreciated. I apologize if this is in the wrong part of the forum but seems like a good place to put it.

Essentially what I would like to do is determine how the various features of a unit effect the rental rate per sq. ft. so that I can assign a value multiplier to those features and estimate a fair price when setting the rate for new units. I would be able to determine a base "featureless" rate per sq. ft. utilizing internal data and competitor rates.

I have quite a bit of internal data to test against but I'm looking to assess the the value (and statistical significance) of features like:

Unit Type
Floor
Climate Control
Burglar Alarm
Unit Square Footage
etc.

In practice I'm thinking it would tell me the premium that a feature (or combination of features) command in a given market so that I can apply the premium to the base rate and generate a relatively accurate rate per sq. ft.

Again, any help that could point me in the right direction would be greatly appreciated.
 

noetsi

Fortran must die
#2
I am not sure what you are asking exactly. If you run a regression model it will tell you how the Y changes with X although that is not the same as the relative impact of any given X on Y which is a complex topic.