Nationwide +1.3 (-6.2 YoY)

wolvoman

IveSeenTheLight wrote: »

It would be roughly the same.
It's a long term trend with over 20 years of data. (Includes 2 previous crashes)

See here for 2007, the trend line didn't go up did it?
http://nationwide.co.uk/hpi/historical/jul_2007.pdf

There's something very wrong with that graph regarding the trend line.

1. It doesn't 'trend down' during the downturns.

2. It shows the trend at 2.7% per annum in Q2 2007. It shows 2.9% in Q2 2009. How on earth could the trend have gone up despite prices having fallen 10% or so in that time?

michaels

I would guess the trend line is the steady growth rate that makes the areas above and below the line equal - answering your second Q the extra time between then and now above the line is what has pulled the line up.

This should also answer your first question - it would be possible to chose a period that would produce a downward trend line using this methodology but not for this period

IveSeenTheLight

Graham_Devon wrote: »

And I never said anything like you are now suggesting.

I just wanted to answer your question of whether I woukld accept when they hit trend they could stagnate.

I answered no after clarifying something. I really don't want arguments today, which is why I set out to clarify before answering. And why I'm clarifying why I was clarifying

There was no mallice, or intent anywhere. I didnt say anything like you are saying above. I was just asking a humble question. I wasn't calling, suggesting, or implying anything you are saying.

I really would like to be able to get passed this "are you suggesting stuff" if it's possible, hence why I have been discussing with you today and taking on your points, and hopefully, can continue to do so. If I say something thats stupid, fair enough, but I don't believe I have suggested anything you are saying above, I believe I just wanted to answer your question.

Hope you take this post as intended

Please understand I am not trying to imply you are saying anything, however by saying you would not accept house prices following the long term trend (remember it's only going to vary marginally, maybe by only roughly 0.5% for the yearly long term trend interest) then you do want house prices to undershoot the trend and therefore as historically, you would likely see a continued boom and bust of HPI with the prices undershoot and then overshooting.

Therefore not long term stabalisation (however it may be unlikely)

IveSeenTheLight

wolvoman wrote: »

There's something very wrong with that graph regarding the trend line.

1. It doesn't 'trend down' during the downturns.

2. It shows the trend at 2.7% per annum in Q2 2007. It shows 2.9% in Q2 2009. How on earth could the trend have gone up despite prices having fallen 10% or so in that time?

1. It would trend downwards if taken over a short period when house prices went down. Statistically however, house prices do always increase over the long term, which this is a reflection of in the long term trend. Take the last two years for example, although house prices dropped on average, they were still higher than the average over the term of the dataset. Even now house prices on average have went back 4-5 years, yet there is still 15 -16 years of data previous to this in which the price was cheaper. therefore over the long term, house prices are still on average higher.

2. Because although prices were dropping it was still above the long term average, thus increasing the long term average. If prices were to undershoot the long term average you would see the percentage drop to reflect the new long term average annual HPI.

So nothing wrong with the graph and hope this clarifies

Dan:_4

wolvoman wrote: »

There's something very wrong with that graph regarding the trend line.

1. It doesn't 'trend down' during the downturns.

2. It shows the trend at 2.7% per annum in Q2 2007. It shows 2.9% in Q2 2009. How on earth could the trend have gone up despite prices having fallen 10% or so in that time?

Because house prices ALWAYS rise LONG TERM - hence the trend line. You can see the downturns where 'Real House Prices' go below the trend line.

julieq

michaels wrote: »

I would guess the trend line is the steady growth rate that makes the areas above and below the line equal - answering your second Q the extra time between then and now above the line is what has pulled the line up.

This should also answer your first question - it would be possible to chose a period that would produce a downward trend line using this methodology but not for this period

I'm obviously not explaining this right

It has nothing to do with areas on each side of the graph.

You average the annual growth rate up to a particular time. In this case it gives 2.9% pa on average at July 2009.

You take the start date arbitrarily and note the average price at that point.

You then apply progressively the average rise (i.e. 2.9%) to the start price and compound it for successive points (i.e. x, x*1.029, x*1.029*1.029, etc). So you have a curve curving up, it can't be any other shape.

Because it's averaged it's difficult to affect it much via short term differences unless they're very big indeed.

Graham_Devon

julieq wrote: »

I'm obviously not explaining this right

It has nothing to do with areas on each side of the graph.

You average the annual growth rate up to a particular time. In this case it gives 2.9% pa on average at July 2009.

You take the start date arbitrarily and note the average price at that point.

You then apply progressively the average rise (i.e. 2.9%) to the start price and compound it for successive points (i.e. x, x*1.029, x*1.029*1.029, etc). So you have a curve curving up, it can't be any other shape.

Because it's averaged it's difficult to affect it much via short term differences unless they're very big indeed.

I think your just complicating it

House prices rise in a boom, trend line sits further up than natural as the boom pulls the line up.

House prices crash in a bust, trend line gets pulled back down to a more natural level, or below natural level.

It's gravity.

Simples.

IveSeenTheLight

Graham_Devon wrote: »

I think your just complicating it

House prices rise in a boom, trend line sits further up than natural as the boom pulls the line up.

House prices crash in a bust, trend line gets pulled back down to a more natural level, or below natural level.

It's gravity.

Simples.

She isn't Graham

to clarify, the trend line averagea out all the booms and the busts to give an average.
When taken over a long period such as the 20 year NW graph, it shows on average how house prices have faired over that period.
Given that the graph includes two previous corrections and almost over the term of a standard mortgage, I believe it is a decent indicator of where house prices on average should be.

I guess you could as well extraploate the trend going forward as an indicator of where house prices will be on average.

Is yearly 2.9% HPI (agreed this has been pulled up by recent years) really unsustainable?

julieq

No, Graham, short term fluctuations won't change it much at all. The more points you're averaging over, the more difficult it is for the last few to affect the average, because you have the accumulated weight of the previous ones.

That's why averages (and rolling averages) are used, because they filter out short term fluctuations. But an exponential trend isn't even a graph of averages, it's a compound curve based on the long term average growth.

Look at it this way, if you have a set of 40 numbers all 3, how much does the 41st number need to change by to get the average to 3.5?

ad9898_3

GSXRCarlos wrote: »

Yahoo have just shouted this out to everyone

here we go again!

Here is my quote from yesterday on the subject..

TBH after Halifax's +2.5% in May followed by -0.5% in June, these figures are utterly meaningless, the only irritating thing is if it's a positive it will be hailed from every media outlet like the second coming of Christ, if it's a negative you wouldn't have a clue unless you looked on Forex for yourself.

As I suspected, remember the 0.5% drop from Halifax last month ???, nope thought not, it was hardly reported. With this kind of bias, sentiment against rising prices doesn't stand a chance.