Back to House Prices… • Our failure to reject the null hypothesis implies that the housing stock has no effect on prices – Note the phrase “cannot reject” • This is not very plausible. Even if it is true maybe it is an effect of the bubble – Prices became divorced from their usual determinants • Re-estimate the model for the pre bubble period and see if there is difference • There seems to be a difference after 1997
17
Embed
Back to House Prices… Our failure to reject the null hypothesis implies that the housing stock has no effect on prices – Note the phrase cannot reject.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Back to House Prices…
• Our failure to reject the null hypothesis implies that the housing stock has no effect on prices– Note the phrase “cannot reject”
• This is not very plausible. Even if it is true maybe it is an effect of the bubble– Prices became divorced from their usual determinants
• Re-estimate the model for the pre bubble period and see if there is difference
• There seems to be a difference after 1997
Structural Break
• This is known as a structural break or a regime shift
• Implies that the coefficients may be different not just the variables
• So the conditional expectation function has a kink
• Can happen at a point in time or for a different group of observations
A Regime Shift
2( | )E Y X
X : slope coefficient: Change in E(Y|X)
for a change in x.
Show three data points for illustration
Y E(Y|X)=b1+b2X
Y1 u1
Y3 u3
Y2 u2
b1 X2 X1 X3 X
010
0000
2000
0030
0000
4000
0050
0000
pric
e of
sec
ond
hand
hou
ses
dubl
in, d
oe
1970 1980 1990 2000 2010year
House Prices
Estimating with Structural Break
• Stata command: regress … if condition
regress price inc_pc hstock_pc if year<=1997
Source | SS df MS Number of obs = 28-------------+------------------------------ F( 2, 25) = 88.31 Model | 1.1008e+10 2 5.5042e+09 Prob > F = 0.0000 Residual | 1.5581e+09 25 62324995.9 R-squared = 0.8760-------------+------------------------------ Adj R-squared = 0.8661 Total | 1.2566e+10 27 465423464 Root MSE = 7894.6