The Illinois Association of Realtors said Monday that the median price it reported for home sales within the city of Chicago was inflated in May and mistakes in its reports may go back more than three years. Errors in the reports can wrongly inflate consumer confidence in a housing market that has been struggling to recover for the past 4 1/2 years. It also can undermine the credibility of the... [more]
The Illinois Association of Realtors said Monday that the median price it reported for home sales within the city of Chicago was inflated in May and mistakes in its reports may go back more than three years.
Errors in the reports can wrongly inflate consumer confidence in a housing market that has been struggling to recover for the past 4 1/2 years. It also can undermine the credibility of the real estate organizations that compile and disseminate the statistics. The Tribune and other media outlets report that data as part of regular coverage of the housing industry because it provides a pulse of the market.
http://www.chicagotribune.com/business/chi-realty-trade-group-overreported-chicago-home-prices-20110711,0,7597129.story[less]
Response by West81st
about 15 years ago
Posts: 5564
Member since: Jan 2008
From +10.3% to -8.3% is quite an adjustment. But hey, medians are tricky.
Ignored comment.
Unhide
Response by sidelinesitter
about 15 years ago
Posts: 1596
Member since: Mar 2009
There are a couple of typos in the thread title, specifically the word 'if' and the question mark
Ignored comment.
Unhide
Response by nyc_sport
about 15 years ago
Posts: 820
Member since: Jan 2009
That article is priceless. Even if one were to assume that the "error" was "inadvertent," the fact that the real estate "professionals" do not even know directional pricing in their own market, even in their own niches, is mind-boggling. Medians are tricky, + and - signs are pretty basic.
Ignored comment.
Unhide
Response by Riversider
about 15 years ago
Posts: 13573
Member since: Apr 2009
Calculated risk has pointed out bad NAR numbers for some time.
At best they are just sloppy. At worst they are doing something intentionally sneaky and self serving.
--------------------
The National Association of Realtors (NAR) responded today to recent stories about the NAR overstating sales. This was in response to Calculated Risk posts in January (here, here and here), and to CoreLogic's release on Tuesday. Barry Ritholtz picked up on the story this morning.
Can you ever really trust anyone who is trying to sell you something and gets paid on commission?
Ignored comment.
Unhide
Response by rb345
about 15 years ago
Posts: 1273
Member since: Jun 2009
You cant extrapolate from that one event: it's Chicago; everything in that city
is corrupt, except for the untreated sewage, which is purer than its politicians.
Ignored comment.
Unhide
Response by Riversider
about 15 years ago
Posts: 13573
Member since: Apr 2009
Realtor defends:
"Median price in the market does not matter at all," said Stacy Karel, an @Properties agent. "It's building or block specific. It's a micromarket. You can't compare east of Western (Avenue) with west of Western." When Rojas has pulled data directly from Midwest Real Estate Data, it shows that even the best city neighborhoods remain depressed.
Ignored comment.
Unhide
Response by urbandigs
about 15 years ago
Posts: 3629
Member since: Jan 2006
thats why Im slowly building http://www.urbandigs.com to get analytics on the manhattan marketplace..its not easy though. For Manhattan, sales data should be adapted to local market characteristics in terms of normalcy of public record filing lags. For Manhattan, this lag is somewhere between 2-4 weeks, osssibly extending to a few months. That means, that median reports will be INCOMPLETE if published in real time. Rather, sales data and sales metrics (absorption rate, months of supply, median, avg sales price, etc), should wait for some kind of completion rate to be achieved - maybe 90%.
In our Manhattan market, I would argue you need to wait 90 days or so to get to that completion percentage. If you publish June's median sales charts/trends on July 1st, maybe you have 50% of actual sales that closed in the month of June. The rest will roll in over time. When completion percentage is closer to 90%, the #s are likely to change drastically..that could mean an initial signal of +5% may in fact be revised to a -5% once all sales are collected and counted.
So, median #s are misleading if published too early. Wait, its better to get that bigger picture using 2x more data
Ignored comment.
Unhide
Response by Riversider
about 15 years ago
Posts: 13573
Member since: Apr 2009
Agree on the publishing date.
Why not use quintiles?
Ignored comment.
Unhide
Response by urbandigs
about 15 years ago
Posts: 3629
Member since: Jan 2006
we can do that...but I dont really see the point in tracking the early holding periods
From +10.3% to -8.3% is quite an adjustment. But hey, medians are tricky.
There are a couple of typos in the thread title, specifically the word 'if' and the question mark
That article is priceless. Even if one were to assume that the "error" was "inadvertent," the fact that the real estate "professionals" do not even know directional pricing in their own market, even in their own niches, is mind-boggling. Medians are tricky, + and - signs are pretty basic.
Calculated risk has pointed out bad NAR numbers for some time.
At best they are just sloppy. At worst they are doing something intentionally sneaky and self serving.
--------------------
The National Association of Realtors (NAR) responded today to recent stories about the NAR overstating sales. This was in response to Calculated Risk posts in January (here, here and here), and to CoreLogic's release on Tuesday. Barry Ritholtz picked up on the story this morning.
http://www.calculatedriskblog.com/2011/02/nar-reponds-to-questions-of-overstating.html
Like we didn't know 99% of brokers were scum.
Can you ever really trust anyone who is trying to sell you something and gets paid on commission?
You cant extrapolate from that one event: it's Chicago; everything in that city
is corrupt, except for the untreated sewage, which is purer than its politicians.
Realtor defends:
"Median price in the market does not matter at all," said Stacy Karel, an @Properties agent. "It's building or block specific. It's a micromarket. You can't compare east of Western (Avenue) with west of Western." When Rojas has pulled data directly from Midwest Real Estate Data, it shows that even the best city neighborhoods remain depressed.
thats why Im slowly building http://www.urbandigs.com to get analytics on the manhattan marketplace..its not easy though. For Manhattan, sales data should be adapted to local market characteristics in terms of normalcy of public record filing lags. For Manhattan, this lag is somewhere between 2-4 weeks, osssibly extending to a few months. That means, that median reports will be INCOMPLETE if published in real time. Rather, sales data and sales metrics (absorption rate, months of supply, median, avg sales price, etc), should wait for some kind of completion rate to be achieved - maybe 90%.
In our Manhattan market, I would argue you need to wait 90 days or so to get to that completion percentage. If you publish June's median sales charts/trends on July 1st, maybe you have 50% of actual sales that closed in the month of June. The rest will roll in over time. When completion percentage is closer to 90%, the #s are likely to change drastically..that could mean an initial signal of +5% may in fact be revised to a -5% once all sales are collected and counted.
So, median #s are misleading if published too early. Wait, its better to get that bigger picture using 2x more data
Agree on the publishing date.
Why not use quintiles?
we can do that...but I dont really see the point in tracking the early holding periods