predicting the housing market with buyer & seller psychology (predictive analytics summit 2011)
DESCRIPTION
Presented by Scott Sambucci (Altos Research) at the IE Group’s Predictive Analytics Summit, "Predicting the Housing Market with Buyer & Seller Psychology" reviews real-time and leading indicators to predict future home prices.TRANSCRIPT
Scott Sambucci, Market Analytics
Altos ResearchFebruary 2011 | San Diego, CA
PREDICTING THE HOUSING MARKET WITH BUYER & SELLER PSYCHOLOGY
PREDICTIVE ANALYTICS INNOVATION SUMMIT
WHO, WHAT, WHERE, WHEN, WHY & HOW
• Altos Research
• Real-time housing market analytics
• 200+ metros, 20,000 zip codes
• Not what happened but what is happening & what will happen
• Wisdom of crowds
• Using Active Market to forecast future market activity
TODAY’S ACTIVE MARKET SIGNALS FUTURE TRANSACTION PRICES
Price reduced$398,000
Inventory 69
Neighbor home listed$409,000
Deal closed$389,000
Transaction Recorded
Buyer financing fails,Property relisted
$394,000
March May July Sept Nov Jan
Closed transaction = 1 data point, months too lateActive Market = 9 months of pricing, price changes, supply and demand, leading indicators
Offer made$391,000
Home listed$429,000
Inventory 49
APPLICATIONS OF REAL-TIME ANALYTICS
• Current Market Conditions & Property Valuation Changes
• Future Housing Market Prices
• Consumer Sentiment
• Financial Market Movements
• Housing Policy & Affordability Implications
3507 ARCLIGHT COURT,FT MYERS, FL
3507 ARCLIGHT COURT, FORT MYERS, FL
SELLER RESPONSE TO MARKET DEMAND:LISTING PRICE HISTORY - 3204 TRAWLER PLACE
MIAMI CONDOS: MARKET IMPROVED IN 2009?
Did values really move higher when
the actives knew the market was weak?
DEVELOPING A FORWARD VALUATION MODEL™:ENSEMBLE OF REGRESSION TREES WITH VARIABLE IMPORTANCE• Non-parametric
• Capture non-linearity & variable interactions
• Embarrassingly parallel
• Train quickly using the cloud
• Bootstrap holdouts
• Use all of our data for training
• Information gain feature selection
• Most important active market statistics
• How entropic is the distribution of the dependent variable with each variable, and then without?
• Modified ARCH model – hybrid of transaction and active market prices
BENCHMARK MODEL RESULTS• Average change in median price over the period was -0.53%
• Second pass, multivariate OLS regression. Modelhad R2 = 29.4%, and mean absolute error (MAE) = 0.60%• This error completely swamps the signal, but why?
• Residuals (errors) show bias and extreme non-linearity:
-2.0%-1.5%-1.0%-0.5%0.0%0.5%1.0%1.5%2.0%2.5%3.0%
ALTOS LPI™ TIME-SERIES RESULTSOut-of-Sample-Out-of-Time results for Mesa, AZ:
11-D
ec
16-D
ec
21-D
ec
26-D
ec
31-D
ec5-J
an10
-Jan15
-Jan20
-Jan25
-Jan30
-Jan4-F
eb9-F
eb
14-Feb
19-Feb
24-Feb
1-Mar
6-Mar
11-M
ar
16-M
ar
21-M
ar
26-M
ar
31-M
ar5-A
pr
10-Apr
15-Apr
20-Apr
25-Apr
30-Apr
5-May
10-M
ay
15-M
ay
20-M
ay
25-M
ay
30-M
ay4-J
un9-J
un14
-Jun
-14.0%
-12.0%
-10.0%
-8.0%
-6.0%
-4.0%
-2.0%
0.0%
Change in Median Price (+3M)
Forecast Change in LPI (+3M)
FVM™ IN-SAMPLE SCATTERPLOT
San Mateo, CA 94401 / Single-family-home / All Quartile / 6M
x-axis is +6M change in LPI, y-axis is forecast +6M change in LPI
-10.0% -8.0% -6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0%
-8.0%
-6.0%
-4.0%
-2.0%
0.0%
2.0%
4.0%
6.0%
8.0%
FVM™ OUT-OF-SAMPLE SCATTERPLOT
San Mateo, CA 94401 / Single-family-home / All Quartile / 6M
x-axis is +6M change in LPI, y-axis is forecast +6M change in LPI
-10.0% -8.0% -6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0% 8.0% 10.0%
-8.0%
-6.0%
-4.0%
-2.0%
0.0%
2.0%
4.0%
6.0%
8.0%
POLICY IMPLICATIONS:2009-2010 FIRST-TIME HOMEBUYER TAX CREDIT
GAMBLING IN LAS VEGAS? SHOCKING…
DECLINING HOUSING DEMAND
MACROECONOMIC INFORMATION GAIN:CONSUMER SENTIMENT & HOUSING
CONSUMER CONFIDENCE & HOME PRICES: 2010
S&P 500 VS HOME PRICES JAN 2008 – NOV 2010
1-Jan
-08
1-Feb
-08
1-Mar-
08
1-Apr-0
8
1-May
-08
1-Jun-08
1-Jul-0
8
1-Aug-08
1-Sep
-08
1-Oct-
08
1-Nov-0
8
1-Dec
-08
1-Jan
-09
1-Feb
-09
1-Mar-
09
1-Apr-0
9
1-May
-09
1-Jun-09
1-Jul-0
9
1-Aug-09
1-Sep
-09
1-Oct-
09
1-Nov-0
9
1-Dec
-09
1-Jan
-10
1-Feb
-10
1-Mar-
10
1-Apr-1
0
1-May
-10
1-Jun-10
1-Jul-1
0
1-Aug-10
1-Sep
-10
1-Oct-
10
1-Nov-1
0500.00
600.00
700.00
800.00
900.00
1,000.00
1,100.00
1,200.00
1,300.00
1,400.00
1,500.00
$325,000
$335,000
$345,000
$355,000
$365,000
$375,000
$385,000
$395,000
$405,000
$415,000
S&P 500 Active Prices
S&P 500 VS. HOME PRICES: MOST RECENT 6 MONTHS
AFFORDABILITY MEASURES: “CAPITA PER INVENTORY”
50 100 150 200 250 300 350 400 450 500 550 $-
$100,000
$200,000
$300,000
$400,000
$500,000
$600,000
$700,000 f(x) = 1408.09308751914 x + 10121.9273527503R² = 0.737013300503559
2011-02-18 50 100 150 200 250 300 350 400
$-
$100,000
$200,000
$300,000
$400,000
$500,000
$600,000
$700,000
$800,000
f(x) = 1642.89086843538 x + 115986.573286277R² = 0.684094088497661
9/26/08
AFFORDABILITY: CAPITA PER INVENTORY
50 100 150 200 250 300 350 400 450 500 550 $-
$100,000
$200,000
$300,000
$400,000
$500,000
$600,000
$700,000 f(x) = 1408.09308751914 x + 10121.9273527503R² = 0.737013300503559
2011-02-18 50 100 150 200 250 300 350 400
$-
$100,000
$200,000
$300,000
$400,000
$500,000
$600,000
$700,000
$800,000
f(x) = 1642.89086843538 x + 115986.573286277R² = 0.684094088497661
9/26/08
Detroit
New York; San Diego
San DiegoNew York
DC
Detroit
DC
Cleveland
Cleveland
Miami
Miami
CONTACT
Scott Sambucci, VP Market Analytics
Altos Research
www.altosresearch.com
(415) 931-7942
http://blog.altosresearch.com
@scottsambucci