Jump to content
House Price Crash Forum

Spline's Hpi Predictor Is Shooting Up


Recommended Posts

Yo know, when predictions atart to go wildy wrong, its usually a sign of crisis, a change in the conditions of the economy.

It appears we are at another crisis.....the next leg down.

How many times do you have to be wrong before it begins to grate a tad? What pray tell makes you think the predictor is wrong?

Apart from looking at that site Spline appears to be predicting 0-3% for the next 5 years. Given inflation in that period could be running at 8%+ it looks a bit hollow. Shooting up certainly isn't what the figure say to me.

Edited by GrillsBears
Link to comment
Share on other sites

How many times do you have to be wrong before it begins to grate a tad? What pray tell makes you think the predictor is wrong?

I looked at a few graphs, saw a wide variation in the lines compared to other times.

this is a sign of change...a change will be required in the model too.

Link to comment
Share on other sites

spline's the only guy who's always called it right...

http://www.houseprices.uk.net/articles/house_price_predictor/

It's going up up and away. I might turn bull. Still can't afford a house though.

ooo another bear turning bull, keep em coming.

the pivotal point though is about 150K, if it falls through that level then its goodnight irene, there is nothing but Air holding them above about 80,000

Link to comment
Share on other sites

My guess is his adjustor to determine the 'neutral level of approvals' (which he presumably introduced to take into account the fact that house prices were rising whilst his predicitve factor of mortgage approvals was so small) will now prove to be his undoing.

I'm willing to take bets that hpi will not be 18% end Q1. Call me 'crazy'.

The neutral level of approvals was there well before the recent rises, and predicted them. I didn't believe them and thought the model had broken down.

Link to comment
Share on other sites

Apart from looking at that site Spline appears to be predicting 0-3% for the next 5 years. Given inflation in that period could be running at 8%+ it looks a bit hollow. Shooting up certainly isn't what the figure say to me.

Where does it say that? You're not looking at the property futures lines are you? They are not his predictions.

Link to comment
Share on other sites

If that is the case then that is very impressive. I've always admired Spline's work by the way in case I give the wrong impression. I suppose I'm a little in the Bloo camp in that I think the model may not keep up with reality here. All that is left to do is wait and see.

I've not had any takers on my bet that it won't hit 18% end Q1 though.

The model clearly over- and under-shoots. But the direction always seems right.

I think the model must be wrong too. But I've been wrong before, and spline (to his immense credit) has always simply let the numbers do the talking, and not commented on whether he thinks the predictions are sensible, or why.

Link to comment
Share on other sites

If that is the case then that is very impressive. I've always admired Spline's work by the way in case I give the wrong impression. I suppose I'm a little in the Bloo camp in that I think the model may not keep up with reality here. All that is left to do is wait and see.

I've not had any takers on my bet that it won't hit 18% end Q1 though.

Models work until they don't.

Poor ones work for a short time. Excellent ones work for a long time. In the end, an aspect of reality that is omitted from the model causes it to fail.

Financial history is littered with examples.

Link to comment
Share on other sites

How many times do you have to be wrong before it begins to grate a tad? What pray tell makes you think the predictor is wrong?

Apart from looking at that site Spline appears to be predicting 0-3% for the next 5 years. Given inflation in that period could be running at 8%+ it looks a bit hollow. Shooting up certainly isn't what the figure say to me.

yup - real-terms falls of 2-5% pa for a decade leads to further 40% from now, real-terms, albeit hidden from the average pleb by inflation

Link to comment
Share on other sites

If that is the case then that is very impressive. I've always admired Spline's work by the way in case I give the wrong impression. I suppose I'm a little in the Bloo camp in that I think the model may not keep up with reality here. All that is left to do is wait and see.

I've not had any takers on my bet that it won't hit 18% end Q1 though.

indeed, Splines work has been impressive.

but models always break down. if they didnt, we'd ALL be milyonairs.

one thing this model wont be able to predict is the interference in the market by A: banks not possessing, B: banks not selling possessed stock.

the market is aslo being interfered with by government aiding normally forced sellers to keep their properties off the market.

I see a time of change.

Link to comment
Share on other sites

ooo another bear turning bull, keep em coming.

the pivotal point though is about 150K, if it falls through that level then its goodnight irene, there is nothing but Air holding them above about 80,000

errr.. there's nothing but air in that statement. Please at least give some sort of explanation for this hypothesis.

Link to comment
Share on other sites

yup - real-terms falls of 2-5% pa for a decade leads to further 40% from now, real-terms, albeit hidden from the average pleb by inflation

1) Real term falls are useless for people holding cash - only nominal falls matter

2) Real term falls but nominal price rises still make the gains significant as most sheeple are leveraged.

Link to comment
Share on other sites

one thing this model wont be able to predict is the interference in the market

From the ever-factual Whackopaedia...

At 1 p.m. on the same day (October 24), several leading Wall Street bankers met to find a solution to the panic and chaos on the trading floor.[12] The meeting included Thomas W. Lamont, acting head of Morgan Bank; Albert Wiggin, head of the Chase National Bank; and Charles E. Mitchell, president of the National City Bank of New York. They chose Richard Whitney, vice president of the Exchange, to act on their behalf. With the bankers' financial resources behind him, Whitney placed a bid to purchase a large block of shares in U.S. Steel at a price well above the current market. As traders watched, Whitney then placed similar bids on other "blue chip" stocks. This tactic was similar to a tactic that ended the Panic of 1907, and succeeded in halting the slide that day.

... and like a day-time infomercial, wait, there's more!...

An interim bottom occurred on November 13 [ ... 1929 ... ] with the Dow closing at 198.60 that day. The market recovered for several months from that point, with the Dow reaching a secondary closing peak (i.e., bear market rally) of 294.07 on April 17, 1930. The market embarked on a steady slide in April 1931 that did not end until 1932 when the Dow closed at 41.22 on July 8, concluding a shattering 89% decline from the peak.

The heavily indebted stand about as much chance of clearing their mortgages as the midwest did of stumping up their margin back on the 28th October, 1929.

2500 on the Dow, sometime this decade.

Edited by ParticleMan
Link to comment
Share on other sites

spline's the only guy who's always called it right...

http://www.houseprices.uk.net/articles/house_price_predictor/

It's going up up and away. I might turn bull. Still can't afford a house though.

I think this is bull. Technical analysis based on empirical evidence of past trends despite the fact that the underlying fundamentals are undergoing a phase change?

I'm not expecting a major crash as a result of the new fundamentals but I think this predictor which derives its value from past correlation is extremely suspect. I certainly wouldn't buy a house on the strength of it, or any other technically based measure to assess future house prices.

Link to comment
Share on other sites

spline's the only guy who's always called it right...

http://www.houseprices.uk.net/articles/house_price_predictor/

It's going up up and away. I might turn bull. Still can't afford a house though.

Hmm, the filter he is using undershot significantly last year. I suspect it will overshoot this year. Looks like a misapplication of an off the shelf tool to me. I'd also like to know whether the filter he is using only includes past data points or future ones if available. Without spending too much time looking at this article, I suspect the latter, which might explain why it is such a good fit for the first part of the graph.

(Kalman filters are a bit of a bugbear of mine, as I've run into them being applied blindly all too often.)

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.




×
×
  • Create New...

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.