Jump to content
House Price Crash Forum
Sign in to follow this  
crashmonitor

The House Price Indicies All Over The Place

Recommended Posts

Maybe they are starting to agree on direction, but we have a 90 grand difference on the actual average price. As pointed out on another thread the land registry is now the propaganda tool of choice for justification of ZIRP and HTB2.

Meanwhile the ones that try and iron out sampling bias such as NOS have us at peak prices ...£247,000

http://www.telegraph.co.uk/finance/personalfinance/houseprices/10403980/Average-UK-house-price-165000-or-252000.html

Share this post


Link to post
Share on other sites

And not one mention of geometric vs. arithmetic average to help explain the gap between ONS and Land Reg indices. Just what do these hacks get paid for?

Edit, only in our sanctified national press could we have an article which sets out to explain and demystify the various property indices which has almost no recourse to the methodologies employed.

e.g in attempting to explain the difference between ONS and Land Reg, the author states:

Another Government-backed organisation, the Office of National Statistics, produces one of the most complex indices which “together with the Land Registry HPI is one of the main indices used by central and local Government”. But why does it come up with such different numbers? The short answer is that ONS statisticians obtain mortgage data from lenders and then “mix-adjust” it to reflect different property types and locations. It is also weighted to reflect volumes of transactions, in a process which is updated annually. There is also a seasonal adjustment.

None of which is also employed by the Land Reg? No mention of repeat sales regression for Land Reg sales either, the core means of generating changes to the index since it was calculated 'for real' in ~April 2000?

Edited by The B.L.T.

Share this post


Link to post
Share on other sites

And not one mention of geometric vs. arithmetic average to help explain the gap between ONS and Land Reg indices. Just what do these hacks get paid for?

Edit, only in our sanctified national press could we have an article which sets out to explain and demystify the various property indices which has almost no recourse to the methodologies employed.

e.g in attempting to explain the difference between ONS and Land Reg, the author states:

None of which is also employed by the Land Reg? No mention of repeat sales regression for Land Reg sales either, the core means of generating changes to the index since it was calculated 'for real' in ~April 2000?

I'm still struggling to see where this 90 grand difference comes from. Arithmetic and Geometric averages will surely have more effect on annual percentage changes than coming up with different average values.

Share this post


Link to post
Share on other sites

I'm still struggling to see where this 90 grand difference comes from. Arithmetic and Geometric averages will surely have more effect on annual percentage changes than coming up with different average values.

The Land Registry takes the geometric mean in April 2000 and adjusts this forward based on the calculated change in house prices using repeat sales regression. If you take the geometric mean today then it's quite a bit higher – around 185K.

Additionally, the ONS index is based on a sample of mortgage completions only (65%-70% of all mortgage sales). If cash sales were included the average would be lower.

In the latest Land Registry dataset (82,000 transactions) the difference between the arithmetic and geometric mean is 59K.

BTW, the ONS aren't happy with the current index, and they are proposing to develop a single definitive house price index that incorporates cash sales and gives regional level data. We should get to hear some more details on this early next year.

Share this post


Link to post
Share on other sites

I'm still struggling to see where this 90 grand difference comes from. Arithmetic and Geometric averages will surely have more effect on annual percentage changes than coming up with different average values.

They do generate surprisingly different results, over £60,000. Have a look here http://www.doogal.co.uk/PropertySales.php where I've generated averages for the raw LR data. The full set of data going back to 1995 is meant to be coming out this month, which I'll put on that page, if my server doesn't melt down...

Share this post


Link to post
Share on other sites

They do generate surprisingly different results, over £60,000. Have a look here http://www.doogal.co...opertySales.php where I've generated averages for the raw LR data. The full set of data going back to 1995 is meant to be coming out this month, which I'll put on that page, if my server doesn't melt down...

The variance is too great for the geometric mean to portray confidently, perhaps they should use the harmonic mean instead?

Share this post


Link to post
Share on other sites

The Guardian did a much better analysis of indices. The Telegraph fails to take into account the main differences and their main shortcomings. The problem with average prices is that one high end sale can distort the number quite significantly. It is rarely, if ever, an accurate measure of price movements. The median is better than the average. It measures the mid point of the price spectrum. It too, however, becomes distorted when price differentials are large, so it rarely gives an accurate picture on a local level. It therefore points more to where (price range) sales are happening. The most accurate measure is an index based on repeat sales. This measures what homes sold for, and then what they resold for. This is what the UK Land Registry uses and is the most accurate. I would be very wary of using any index belonging to a lender or agency as they invariably paint a picture that serves their purposes. That the indices vary so much is ipso facto a clear indication they are not representative. The level of transparency in the housing market is appalling and this contributes considerably to distortions. It is about time the consumer and the media demanded access to information to help make better choices.

FreeTrader has spent plenty of time highlighting shortcomings in the LR release, from decimal points slipping on sale prices(the resulting loony price was subsequently used as a headline for the press release without even a cursory check) to high value commercial sales turning up in the index.

It's always a challenge to get the data up to a decent standard but some of the LR failings have shown that it's not to be assumed that even the most basic checks are carried out.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

  • The Prime Minister stated that there were three Brexit options available to the UK:   206 members have voted

    1. 1. Which of the Prime Minister's options would you choose?


      • Leave with the negotiated deal
      • Remain
      • Leave with no deal

    Please sign in or register to vote in this poll. View topic


×

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.