Question about Daft.ie Rental Price Report Average Monthly Rents and Yields

Investor

Registered User
Messages
72
Does anyone know whether the average prices and yields in the Daft Rental Price Reports are means or medians? I haven’t been able to find any information on how the numbers are arrived at either in the reports or via a cursory online search.
 
There is a reference to a methodology document in previous reports but it seems gone now, even with Wayback Machine.

My guess is that with the statistical methdology employed you will get means, not medians.

Rents should display a rightward skew of course, so that mean will be usually higher than median.



Edit - I was mixed up, sorry.
 
Last edited:
Why a rightward skew?
Properties have to have minimum standards which puts a floor on prices. Meanwhile very high end properties exist.

Also, the wage distribution has a rightward skew so it's not unreasonable to expect that a product that takes up a big share of income would do as well.
 
There is a reference to a methodology document in previous reports but it seems gone now, even with Wayback Machine.

My guess is that with the statistical methdology employed you will get means, not medians.

Rents should display a rightward skew of course, so that median will be usually higher than mean.
Thanks for your reply. I'd also guess that it's the mean. I appreciate that rent distributions would be right-skewed; I take it that you meant to say that the mean would thus be higher than the median, as jpd points out is the case for income distributions.
 
Thanks for your reply. I'd also guess that it's the mean. I appreciate that rent distributions would be right-skewed; I take it that you meant to say that the mean would thus be higher than the median, as jpd points out is the case for income distributions.
Indeed. I was mixed up and have edited the post.

@jpd - indeed, higher income more likely to own. But the income distribution of the renting population will still be right-skewed too I would think.
 
I emailed the author who sent me this reply

In short, the prices (sale or rental) are from a method called hedonic regressions, which is used by price index compilers all over the world to try and estimate the like-for-like change in prices over time of things like houses, cars, mobile phone bills, etc. In short, each property is broken down into its constituent parts (in particular when it's listed, its size in bedrooms and bathrooms, its type, and its location as defined by a series of over 300 micro-markets) and those components are then priced by the method. So the price changes over time are not being driven by the mix of properties changing (e.g. more 1-2 beds now than last year, or more rural than urban). We top this up with a technique called a Cooks Distance filter, which means that any listings that have too much of an effect on the results, because they are outliers effectively, are omitted from the analysis.
 
OK, hedonic regression is complicated maths to find what people are paying for and Cooks Distance is a way of removing outliers but it is arbitrary.

Does that mean they can can come up with whatever answer they want? well maybe not exactly but ...
 
@Brendan Burgess

Still doesn't tell us if he is generating median or mean property type values!

@jpd I've be reading Daft reports since 2008 and watching the local rental market closely too. The Daft numbers have always smelt right
 
They are not the median as that is just the mid-point rent
They are a sort of mean but not the one taught in secondary school - I guess "average" is a better description as using "mean" is quite restrictive and would imply the usual calculation method was used.

To be fair to Daft - the say "average" and are probably consistent in the calculation used each period
 
Still doesn't tell us if he is generating median or mean property type values!

You can figure it out from the description.

If the distribution is 4,5,6,7 and 20 , the median would be 6.
If the distribution is 4,5,6,7 and 200, the median would also be 6.
If they were using the median, there would be no need for a "a Cooks Distance filter, which means that any listings that have too much of an effect on the results, because they are outliers effectively, are omitted from the analysis."

In general, you should assume that when people say "average" they mean the mean, unless they say otherwise.

Brendan
 
I think you're right, yes.

In general, you should assume that when people say "average" they mean the mean, unless they say otherwise.

Yes, I agree. The median is one measure of the average though (so is the mode).

But when you say "average" in common usage you should be talking about the mean or specify otherwise.
 
Back
Top