|
Louis Christopher SQM cocks up his attack on RP Data
|
|
Topic Started: 4 Jul 2012, 10:14 AM (2,842 Views)
|
|
Strindberg
|
4 Jul 2012, 10:14 AM
Post #1
|
- Posts:
- 3,919
- Group:
- Diamond Member
- Member
- #23
- Joined:
- 02/11/2010
|
Louis Christopher has been slagging off the RP Data figures over the last couple of days. He has extended his attack today in his newsletter (sent by email).
He writes the following:
He accuses RP Data of "some truly questionable reads". The only questionable read that I see in the above is Louis' claim that according to the table Sydney fell nearly one percent. The table doesn't say that at all. The one day change is an index change - not a percentage change. Louis doesn't understand the table.
If you're going to attack others it's best not to cock it up. I expect Louis will have to reissue his newsletter.
|
Housing costs to Income broadly unchanged since 1994 - re-ratified here
The People of Australia have the highest median wealth in the World
2002-2012 10 year house price growth the SLOWEST since 1952-1962
1990-2010 20 year house price growth the SLOWEST since 1950-1970
CHRIS BECKER NOW NEUTERED
"There are two kinds of people in this world: ones that fiddle around wondering whether a thing's right or wrong and guys like us." (Hugo to Gagin in Ride the Pink Horse)
|
| |
|
Black_Dragon
|
4 Jul 2012, 10:25 AM
Post #2
|
- Posts:
- 238
- Group:
- Gold Member
- Member
- #298
- Joined:
- 06/02/2011
|
- Strindberg
- 4 Jul 2012, 10:14 AM
Louis Christopher has been slagging off the RP Data figures over the last couple of days. He has extended his attack today in his newsletter (sent by email). He writes the following: He accuses RP Data of "some truly questionable reads". The only questionable read that I see in the above is Louis' claim that according to the table Sydney fell nearly one percent. The table doesn't say that at all. The one day change is an index change - not a percentage change. Louis doesn't understand the table. If you're going to attack others it's best not to cock it up. I expect Louis will have to reissue his newsletter.
Yep its an error and a pretty minor one at that. The point still holds. They do record volatility on their index on a day to day basis.
BTW there is no "attack" or "war" going on here. We simply dispute their index reads and in particular what they have been stating for Melbourne.
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
|
|
"No sympathy for the devil; keep that in mind. Buy the ticket, take the ride...and if it occasionally gets a little heavier than what you had in mind, well...maybe chalk it off to forced conscious expansion: Tune in, freak out, get beaten."My Webpage
|
| |
|
kennyjaiz
|
4 Jul 2012, 10:52 AM
Post #3
|
- Posts:
- 1,155
- Group:
- Platinum Member
- Member
- #11
- Joined:
- 29/10/2010
|
- Black_Dragon
- 4 Jul 2012, 10:25 AM
Yep its an error and a pretty minor one at that. The point still holds. They do record volatility on their index on a day to day basis.
BTW there is no "attack" or "war" going on here. We simply dispute their index reads and in particular what they have been stating for Melbourne.
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
While I do hold my own scepticism with regard to the validity of a "daily value index" of any commodity, it is expected that such measure would display relatively more inherent volatility.
I'm not sure what you mean by:
- Quote:
-
do you believe Melbourne house prices have been rising this year? Well do you?
According to the screen capture, the index indicated an overall 6.62% drop over the past 12 months in Melbourne (not an increase). I have no doubt that daily Melbourne property prices (similar to petrol and bananas) will have fluctuations, displaying higher or lower prices, compared to the previous day. Consider the weekend of the auction of "The Block" - I would imagine the South Melbourne property prices would reflect a substantial increase based on the available sales data (i.e. house prices rose on the day).
|
|
|
| |
|
miw
|
4 Jul 2012, 11:09 AM
Post #4
|
- Posts:
- 3,925
- Group:
- Diamond Member
- Member
- #1,917
- Joined:
- 22/04/2012
|
- Black_Dragon
- 4 Jul 2012, 10:25 AM
Yep its an error and a pretty minor one at that. The point still holds. They do record volatility on their index on a day to day basis.
BTW there is no "attack" or "war" going on here. We simply dispute their index reads and in particular what they have been stating for Melbourne.
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
Well, no I don't. But then the index didn't say it was then and it doesn't say that now. In fact it also says the Melbourne prices have fallen between May 2 and now, so the index fairly unambiguously indicates that Melbourne prices are in a downtrend.
Just because an index is falling does not mean it will never have an uptick. In fact you wouldn't be expecting to see too many consecutive ticks in the same direction unless the situation was extraordinary.
If you really want to do an investigation for potential bogosity, have a look at Brisbane/Gold Coast for the last 2 weeks. Brisbane was moving up quite strongly until 25th June with about a 5:2 ratio between upticks and downticks, but all of a sudden they have 10 consecutive downticks which for any high-frequency index is pretty out of the ordinary.
My suspicion is that it is related to the change of stamp duty on July 1 leading to a very small number of recent transactions coming into the data being relatively swamped by old May data coming in because anyone buying a PPOR and having a contract dated in the second half of June would have to be crazy or asleep at the wheel.
If this is the case, we should see the Brisbane index snap back to mid-June levels in fairly short order. But it would point up a problem you can have with the index when the temporal distribution of incoming data changes.
Remember they are working from contract prices that get into the system anytime from a few days after signing of the contract to some months afterwards because some get reported by the real-estate agents, but some don't arrive until they show up in the government databases. You wouldn't expect any effect from an interest rate cut to show up in the index for at least a couple of weeks after the cut.
|
Last time I followed my dreams it led me to a casino, a bar, a gun shop, then a bank. Never again!
AREPS™
|
| |
|
Gossamer
|
4 Jul 2012, 11:13 AM
Post #5
|
- Posts:
- 614
- Group:
- Gold Member
- Member
- #1,304
- Joined:
- 22/10/2011
|
Strindberg's vehement attacks of anybody who portrays the RE industry in a negative light are becoming tiresome. It is clear too see that this person is somehow involved in th RE industry - Strindberg never attacks anybody who potrays RE in a positive light. Therefore any posts by this person should be treated with caution due to their inability to provide a balanced opinion.
|
|
Common sense is a curse - those who have it need to suffer dealing with those who don't have it.
|
| |
|
Black_Dragon
|
4 Jul 2012, 11:21 AM
Post #6
|
- Posts:
- 238
- Group:
- Gold Member
- Member
- #298
- Joined:
- 06/02/2011
|
- miw
- 4 Jul 2012, 11:09 AM
- Black_Dragon
- 4 Jul 2012, 10:25 AM
Yep its an error and a pretty minor one at that. The point still holds. They do record volatility on their index on a day to day basis.
BTW there is no "attack" or "war" going on here. We simply dispute their index reads and in particular what they have been stating for Melbourne.
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
Well, no I don't. But then the index didn't say it was then and it doesn't say that now. In fact it also says the Melbourne prices have fallen between May 2 and now, so the index fairly unambiguously indicates that Melbourne prices are in a downtrend. Just because an index is falling does not mean it will never have an uptick. In fact you wouldn't be expecting to see too many consecutive ticks in the same direction unless the situation was extraordinary. If you really want to do an investigation for potential bogosity, have a look at Brisbane/Gold Coast for the last 2 weeks. Brisbane was moving up quite strongly until 25th June with about a 5:2 ratio between upticks and downticks, but all of a sudden they have 10 consecutive downticks which for any high-frequency index is pretty out of the ordinary. My suspicion is that it is related to the change of stamp duty on July 1 leading to a very small number of recent transactions coming into the data being relatively swamped by old May data coming in because anyone buying a PPOR and having a contract dated in the second half of June would have to be crazy or asleep at the wheel. If this is the case, we should see the Brisbane index snap back to mid-June levels in fairly short order. But it would point up a problem you can have with the index when the temporal distribution of incoming data changes. Remember they are working from contract prices that get into the system anytime from a few days after signing of the contract to some months afterwards because some get reported by the real-estate agents, but some don't arrive until they show up in the government databases. You wouldn't expect any effect from an interest rate cut to show up in the index for at least a couple of weeks after the cut.
All fair points.
I dont dispute you can have up-ticks in an index when there is a clear down trend. But I do dispute in this case whether the index concerned is correctly capturing the various movements as I do have concerns surrounding how that index is collated and calculated. In the case of Melbourne, I struggle to see how one could record what are fairly substantial upticks of 1.8% and 1% respectively when there appears to be no demand or supply indicators suggesting a real change in demand/supply to cause such an increase in price.
What makes this worse of course is that the authors of the index in question DO think the upticks mean something as illustrated here in March:
http://www.propertyobserver.com.au/residential/flat-lining-australian-housing-prices-return-to-early-2010-levels-christopher-joye/2012032754024
And here just a couple of days back where Chris Joye has suggested dwelling prices "surged"
http://www.propertyobserver.com.au/residential/recovering-house-prices-across-most-australian-capitals-buck-the-typically-weak-june-trend-christopher-joye/2012070255373
|
|
"No sympathy for the devil; keep that in mind. Buy the ticket, take the ride...and if it occasionally gets a little heavier than what you had in mind, well...maybe chalk it off to forced conscious expansion: Tune in, freak out, get beaten."My Webpage
|
| |
|
Strindberg
|
4 Jul 2012, 11:23 AM
Post #7
|
- Posts:
- 3,919
- Group:
- Diamond Member
- Member
- #23
- Joined:
- 02/11/2010
|
- Black_Dragon
- 4 Jul 2012, 10:25 AM
Yep its an error and a pretty minor one at that. The point still holds. They do record volatility on their index on a day to day basis.
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
No, it isn't minor. It concerns the only example of evidence you provided of your "questionable readings" claim.
If today, or yesterday, you have misread the daily change as a percentage then it seems likely you have always been misreading that daily change as a percentage and that fully explains your concern about volatility on their daily index. The daily changes are small and the chart shows no great unexpected volatility.
- Quote:
-
BTW there is no "attack" or "war" going on here. We simply dispute their index reads and in particular what they have been stating for Melbourne.
It certainly comes over as an attack on the RD Data methodology and your comments have been spread all over the media.
- Quote:
-
Sir, do you believe Melbourne house prices have been rising this year? Well do you?
I don't collect or process the data so my view has no importance. However, it appears that all data providers including RP Data agree that Melbourne house prices have fallen this year.
|
Housing costs to Income broadly unchanged since 1994 - re-ratified here
The People of Australia have the highest median wealth in the World
2002-2012 10 year house price growth the SLOWEST since 1952-1962
1990-2010 20 year house price growth the SLOWEST since 1950-1970
CHRIS BECKER NOW NEUTERED
"There are two kinds of people in this world: ones that fiddle around wondering whether a thing's right or wrong and guys like us." (Hugo to Gagin in Ride the Pink Horse)
|
| |
|
kennyjaiz
|
4 Jul 2012, 11:33 AM
Post #8
|
- Posts:
- 1,155
- Group:
- Platinum Member
- Member
- #11
- Joined:
- 29/10/2010
|
- Gossamer
- 4 Jul 2012, 11:13 AM
Strindberg's vehement attacks of anybody who portrays the RE industry in a negative light are becoming tiresome. It is clear too see that this person is somehow involved in th RE industry - Strindberg never attacks anybody who potrays RE in a positive light. Therefore any posts by this person should be treated with caution due to their inability to provide a balanced opinion.
I respect anyone's right to express their opinions in a tolerable manner. Instead of questioning Stringberg's motivation, wouldn't it be more constructive to address the arguments/points he made? If it is unbalanced, it wouldn't too difficult to rebut.
I don't always agree with Strindberg, but I find him to be intelligent and insightful. Some of his analyses are very revealing. You are certainly welcome to ignore it just because it's tiresome, but I would rather consider it.
|
|
|
| |
|
kennyjaiz
|
4 Jul 2012, 11:37 AM
Post #9
|
- Posts:
- 1,155
- Group:
- Platinum Member
- Member
- #11
- Joined:
- 29/10/2010
|
- Black_Dragon
- 4 Jul 2012, 11:21 AM
Which measure are you referring to with the uptick of "1.8%" and "1%"?
|
|
|
| |
|
Black_Dragon
|
4 Jul 2012, 11:47 AM
Post #10
|
- Posts:
- 238
- Group:
- Gold Member
- Member
- #298
- Joined:
- 06/02/2011
|
- kennyjaiz
- 4 Jul 2012, 11:37 AM
- Black_Dragon
- 4 Jul 2012, 11:21 AM
Which measure are you referring to with the uptick of "1.8%" and "1%"?
RP recorded a rise of 1.8% for Melbourne for the month of February and 1% for June
|
|
"No sympathy for the devil; keep that in mind. Buy the ticket, take the ride...and if it occasionally gets a little heavier than what you had in mind, well...maybe chalk it off to forced conscious expansion: Tune in, freak out, get beaten."My Webpage
|
| |
|
miw
|
4 Jul 2012, 11:51 AM
Post #11
|
- Posts:
- 3,925
- Group:
- Diamond Member
- Member
- #1,917
- Joined:
- 22/04/2012
|
- Black_Dragon
- 4 Jul 2012, 11:21 AM
a day's movement will be influenced by whatever data came in the day before. So you could quite possibly have a strong uptick in the middle of a strong downtrend.
I reckon anyone who tries to get any real meaning out of a single tick is reading way too much into what is essentially noise. The RPData index is heavily smoothed, so perhaps a week would start to be significant of something.
Hell. I don't take too much notice of a month's movement in real estate. 3 months moving in the same direction and you can probably say you have a short-term trend.
Edited by miw, 4 Jul 2012, 11:51 AM. |
Last time I followed my dreams it led me to a casino, a bar, a gun shop, then a bank. Never again!
AREPS™
|
| |
|
Shadow
|
4 Jul 2012, 12:07 PM
Post #12
|
Evil Zealot Specufestor
- Posts:
- 6,013
- Group:
- Diamond Member
- Member
- #22
- Joined:
- 02/11/2010
|
- Black_Dragon
- 4 Jul 2012, 11:47 AM
RP recorded a rise of 1.8% for Melbourne for the month of February and 1% for June
A 1% monthly movement up here or there is just noise and/or seasonality.
As MIW says, wait until you see a few consecutive months in the same direction before considering it means anything.
Melbourne is still on a downtrend for the year to date. All the other cities shown are basically flat, with no clear direction either way.
Edited by Shadow, 4 Jul 2012, 12:08 PM. |
Shadow's Blog - The Australian Housing Market
1 - Debunking Demographia. Demographia Survey Debunked. Australian housing is not particularly unaffordable by global standards.
2 - USA, Ireland, UK, Spain and Japan Property Bubbles versus Australia. All confirmed property bubbles had one thing in common... a particular house price/income ratio pattern.
3 - Banks can't margin call on residential property unless borrower defaults, because residential property loans are regulated by the NCCP Act 2009.
4 - Housing is the second highest taxed sector of the Australian Economy. Renters don't pay their fair share of tax, and are subsidised by high taxes incurred by homeowners.
5 - Epic Fail! Steve Keen's Bad Calls and Predictions.
Parse: A rep's spare spear pares pears, reaps as per AREPS.
|
| |
|
Alec Smart
|
4 Jul 2012, 12:15 PM
Post #13
|
- Posts:
- 381
- Group:
- Gold Member
- Member
- #1,579
- Joined:
- 10/01/2012
|
- Strindberg
- 4 Jul 2012, 10:14 AM
Louis Christopher has been slagging off the RP Data figures over the last couple of days. He has extended his attack today in his newsletter (sent by email).
He writes the following:
He accuses RP Data of "some truly questionable reads".
The RP Data-Rismark Index is composed of sales data provided by the Valuer Generals Office in each state and sales data provided by real estate selling agents.
I know from experience as a RP Data private subscriber that the database is not always up to date and is sometimes incorrect.
Firstly I am aware that at times RP Data do not use a 100% sample of all sales data. It is a part sample only e.g. 50% to 70% of all sales. Therefore the RP Data-Rismark Index would be skewed.
Secondly some sales data input by real estate agents is incorrect and this can skew any reports.
Thirdly on a week to week basis the RP Data-Rismark Index quality is only as good as a 50%-70% sample and if the sales data is 100% accurate.
Fourthly on a week to week basis the RP Data-Rismark Index quality will be influenced by the availability of sales data from the VGO and subsequent input of this sales data. e.g. the contract between RP Data-Rismark and the VGO in each state may only allow for data import every 2 weeks or 4 weeks. All state based VGO's may have different sales data provision frequencies. For example RP Data-Rismark may receive property sales data from WA, NT and QLD this last fortnight. Then all other states in the next fortnight or each month. So the RP Data-Rismark Index so called spike may be reflecting a spike in WA, NT and QLD and not all of Australia across the board.
Food for thought.
How accurate and reliable is the RP Data-Rismark Index???
Let's not forget that RP Data and Rismark International may have vested commercial interests to influence the Australian property market. See the following products and services:
1. RP Data-Rismark Indices http://www.rismark.com.au/indicies.html
1. Rismark International Questioned. See the Choice article on risky home loans and the RISMARK/ADELAIDE BANK Equity Finance Mortgage (EFM) http://www.choice.com.au/reviews-and-tests/money/borrowing/your-mortgage/risky-home-loans/page/shared%20equity%20mortgage.aspx
2. RP Data customers are real estate service providers http://www.rpdata.com/solutions/property_professionals.html
2. Rismark partners are Macquarie Group, Bendigo & Adelaide Bank Limited, QBE Lenders’ Mortgage Insurance Limited, RP Data Ltd and others. http://www.rismark.com.au/partners/index.html
2. Paid commercially biased editorial. Check out the CHOICE article here http://www.choice.com.au/reviews-and-tests/money/investing/property/real-estate-puffery.aspx- The relationship between media and real estate is apparently so cosy in some instances that big spenders are kept happy at all costs.
|
|
|
| |
|
kennyjaiz
|
4 Jul 2012, 12:21 PM
Post #14
|
- Posts:
- 1,155
- Group:
- Platinum Member
- Member
- #11
- Joined:
- 29/10/2010
|
- Black_Dragon
- 4 Jul 2012, 11:47 AM
RP recorded a rise of 1.8% for Melbourne for the month of February and 1% for June
Thanks. RPdata's back series do reflect these "uptick". And I get that people putting a positive spin on these measures to paint a rosier picture than reality, is a bit frustrating. But assuming the daily indices are "accurate", the interpretation present by Chris is pretty neutral. Same way you would interpret SQM's substantial increase in stock on market in Melbourne - that is not good news for property prices - which may paint a worse picture than reality.
In terms of the methodology of the data collation and calculation, I don't have the expertise to judge, nor do I have access to it.
But if you determine the convergent validity (i.e. compare) with another indices:
http://www.residex.com.au/indices?state=VIC&dwelling=U#indices
Despite a slight time discrepancy, the Residex data display a similar trend line with an attenuated level of fluctuation as RPdata. Residex data seem to indicate a 0.8% increase for houses, and 1.93% increase for units for the month of March. So is it inconceivable for RPdata to record a 1.8% increase for Feb... you be the judge. Question is, can you draw meaningful conclusions from these. I would think not.
|
|
|
| |
|
Shadow
|
4 Jul 2012, 12:26 PM
Post #15
|
Evil Zealot Specufestor
- Posts:
- 6,013
- Group:
- Diamond Member
- Member
- #22
- Joined:
- 02/11/2010
|
- Alec Smart
- 4 Jul 2012, 12:15 PM
Firstly I am aware that at times RP Data do not use a 100% sample of all sales data. It is a part sample only e.g. 50% to 70% of all sales. Therefore the RP Data-Rismark Index would be skewed.
Secondly some sales data input by real estate agents is incorrect and this can skew any reports.
Thirdly on a week to week basis the RP Data-Rismark Index quality is only as good as a 50%-70% sample and if the sales data is 100% accurate.
Fourthly on a week to week basis the RP Data-Rismark Index quality will be influenced by the availability of sales data from the VG
It is rare for any empirical study to include 100% of the data. The ABS employment numbers for example are based on samples.
Large sample sizes lead to increased precision, and a sample of 50-70% of sales would be huge, and would certainly provide a reasonably precise determination.
Your suggestion that some missing/incorrect data would 'skew' the results would only apply if that missing/incorrect data was itself skewed - for example, if during one month, all homes with less than three bedrooms were missing/incorrect, then this would certainly skew the results.
However, one would normally expect the missing/incorrect data be fairly random, in which case the results wouldn't be skewed at all.
http://en.wikipedia.org/wiki/Sample_size_determination
http://en.wikipedia.org/wiki/Statistical_significance
|
Shadow's Blog - The Australian Housing Market
1 - Debunking Demographia. Demographia Survey Debunked. Australian housing is not particularly unaffordable by global standards.
2 - USA, Ireland, UK, Spain and Japan Property Bubbles versus Australia. All confirmed property bubbles had one thing in common... a particular house price/income ratio pattern.
3 - Banks can't margin call on residential property unless borrower defaults, because residential property loans are regulated by the NCCP Act 2009.
4 - Housing is the second highest taxed sector of the Australian Economy. Renters don't pay their fair share of tax, and are subsidised by high taxes incurred by homeowners.
5 - Epic Fail! Steve Keen's Bad Calls and Predictions.
Parse: A rep's spare spear pares pears, reaps as per AREPS.
|
| |
| 1 user reading this topic (1 Guest and 0 Anonymous)
|
5:30 AM May 25