Date post: | 20-Dec-2015 |
Category: |
Documents |
View: | 213 times |
Download: | 0 times |
What & WHYWhat & WHY
Our goal was to discover the determents of rising home prices and to identify any anomies in historic housing prices.
To figure out if current housing market is over priced – if there is a real estate bubble.
HOWHOW
1) We collected average home prices in the US: on a monthly basis (1975-2002).
2) Then we gathered data we thought would be good determents of home prices
3) Set up a model and ran a regression
4) Modified our model
5) Interpreted the results
Exploratory Data Exploratory Data AnalysisAnalysis
Variables:
Mortgage rates, unemployment rates,CPI, PPI, S&P Index (alterative INV),
and income per capita.
Sources:
Economagic and the St. Louis Fed.
-20000
-10000
0
10000
20000
30000
0
50000
100000
150000
200000
250000
76 78 80 82 84 86 88 90 92 94 96 98 00 02
Residual Actual Fitted
Dependent Variable: AVGHOMESALES
Method: Least Squares
Date: 11/20/02 Time: 18:16
Sample: 1975:01 2002:07
Included observations: 331
Variable Coefficient Std. Error t-Statistic Prob.
INCPERCAP 7.069048 0.021993 321.4248 0.0000
R-squared 0.976423 Mean dependent var 126863.7
Adjusted R-squared 0.976423 S.D. dependent var 50204.56
S.E. of regression 7708.746 Akaike info criterion 20.74112
Sum squared resid 1.96E+10 Schwarz criterion 20.75260
Log likelihood -3431.655 Durbin-Watson stat 0.383130
Dependent Variable: AVGHOMESALES
Method: Least Squares
Date: 11/20/02 Time: 18:03
Sample: 1975:01 2002:07
Included observations: 331
Variable Coefficient Std. Error t-Statistic Prob.
CPI -2620.066 479.5200 -5.463935 0.0000
PPI 1900.434 253.7132 7.490481 0.0000
UNEMP_RATE -1308.266 486.8899 -2.686985 0.0076
INCPERCAP 8.775442 0.947094 9.265654 0.0000
MRTG_RATE -657.8159 311.1963 -2.113829 0.0353
MONTHS 385.3148 172.7806 2.230081 0.0264
C 30352.78 12264.19 2.474911 0.0138
R-squared 0.984578 Mean dependent var 126863.7
Adjusted R-squared 0.984292 S.D. dependent var 50204.56
S.E. of regression 6292.141 Akaike info criterion 20.35291
Sum squared resid 1.28E+10 Schwarz criterion 20.43332
Log likelihood -3361.407 F-statistic 3447.484
Durbin-Watson stat 0.615215 Prob(F-statistic) 0.000000
Dependent Variable: AVGHOMESALES
Method: Least Squares
Date: 11/20/02 Time: 18:39
Sample: 1975:01 2002:07
Included observations: 331
Variable Coefficient Std. Error t-Statistic Prob.
MONTHS 517.8749 4.625557 111.9595 0.0000
C 40896.51 885.9602 46.16066 0.0000
R-squared 0.974425 Mean dependent var 126863.7
Adjusted R-squared 0.974347 S.D. dependent var 50204.56
S.E. of regression 8041.062 Akaike info criterion 20.82853
Sum squared resid 2.13E+10 Schwarz criterion 20.85151
Log likelihood -3445.122 F-statistic 12534.92
Durbin-Watson stat 0.341824 Prob(F-statistic) 0.000000
Avg Home Price over Time
y = 518.47x + 40830R2 = 0.9749
0
50,000
100,000
150,000
200,000
250,000
0 50 100 150 200 250 300 350
Months starting at 1975
Mea
n H
ome
Pric
e
Further Analysis Further Analysis
Changes in income per capita have no effect on changes in mean home prices
This is also true for changes in mortgage, unemployment rates, S&P and CPI.
change income per cap vs change home price
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
-0.01 -0.005 0 0.005 0.01 0.015 0.02 0.025 0.03
% change income per capita
% c
hang
e m
ean
hom
e pr
ice
change morgage rate vs change mean ave home price
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
-0.2 -0.15 -0.1 -0.05 0 0.05 0.1 0.15 0.2
% change morgage rate
% c
hang
e m
ean
hom
e pr
ice
change unemployment vs change home price
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
-0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1
% change unemployment rate
% c
hang
e m
ean
hom
e pr
ice
S&P Index vs Avg Home Price
y = 42345Ln(x) - 146309R2 = 0.9608
10,000
100,000
1,000,000
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
S&P Index
Mea
n H
ome
Pric
e
ConclusionsConclusions
1) real estate prices move in long-term cycles
2) time is most significant variable; it that helps explain price increases:
Avg Home Price over Time
y = 518.47x + 40830R2 = 0.9749
0
50,000
100,000
150,000
200,000
250,000
0 50 100 150 200 250 300 350
Months starting at 1975
Mea
n H
ome
Pric
e
Center for Economic and Policy Research
-in the last 7 years, home sale prices have increased nearly 30 percent more than the overall rate of inflation
-there is no obvious explanation for a sudden increase in relative demand for housing which could explain the price rise
- the only plausible explanation for sudden surge in home prices is the existance of a housing bubble
-major factor driving housing sales is the expectation that housing prices will be higher in the future
- the collapse of the bubble will lead to a loss of between $1.3 trillion and $2.6 trillion of housing wealth