The MSM has duly regurgitated HM Land Registry's announcement that "house prices in England and Wales edged ahead by just 0.1% during the 12 months to the end of June"
*yawn*
As explained before, HMLR's figure for the y-o-y change for any month (in this case for June, published end July) more or less tracks Nationwide's figure for two months previously (in this case April, published end April).
Nationwide reported a y-o-y fall of 1% for April 2008, so while HMLR's figure may be more scientifically accurate, it is woefully out of date.
Tuesday, 29 July 2008
HM Land Registry versus Nationwide House Price Indices (2)
Subscribe to:
Post Comments (Atom)
2 comments:
Hello. I have looked at the algorithm used by the company that calculates HMLR's index. In my opinion it is not more scientifically valid. HMLR do not adjust for mix of housing. They fit a scaling factor to the change in prices of all sold properties compared to their previous sale. If Yorkshire (say) has a lot of sales then it will get a disproportionate influence on their final figure. In other words their index is biased towards regions of the UK where the market is still (relatively) brisk, and therefore has not dropped very much.
Sure, other HPC'ers have linked to methodology and it's largely over my head, but as an average for the UK or even a region is subject to a wide margin of error, it's the overall trend that matters, and, as far as the difference between DCLG/HMLR's figures vs Haliwide's figures go, it's largely a timing thing.
Post a Comment