ECON301_Handout_12_1213_02

# Metu department of economics instructor dr ozan

This preview shows pages 7–10. Sign up to view the full content.

METU - Department of Economics Instructor: Dr. Ozan ERUYGUR e-mail: [email protected] Lecture Notes 7 Y t = 103.7+3.75 X t1 +10.51 X t2 - 8.1 D t2 - 0.85 D t2 X t1 +2.79 D t2 X t2 2 nd Param. If we illustrate the second parametrization, we can write as follows: Y t = B 01 +B 11 X t1+ B 21 X t2 +(B 02 - B 01 ) D t2 +(B 12 - B 11 )D t2 X t1 +(B 22 - B 21 )D t2 X t2 +u t 1.subsample is base 02 12 22 (D t1 = 1 - D t2 ) Intercept Slope Slope dummy dummy dummy b) First parametrization Y t = B 01 +B 11 X t1 +B 21 X t2 +u t 1st Subsample Y t = B 02 +B 12 X t1 +B 22 X t2 +u t 2nd Subsample Y t = B 01 D t1 +B 11 D t1 X t1 +B 21 D t1 X t2 +B 02 D t2 +B 12 D t2 X t1 +B 22 D t2 X t2 +u t According to second parametrization: SU=S1+S2=7262+1131=8393 Y t = B 01 +B 11 X t1+ B 21 X t2 +(B 02 - B 01 ) D t2 +(B 12 - B 11 )D t2 X t1 +(B 22 - B 21 )D t2 X t2 +u t Y t = B 01 +B 11 X t1+ B 21 X t2 + 02 D t2 + 12 D t2 X t1 + 22 D t2 X t2 +u t (2nd Param.) Joint Test H 0 : 02 = 12 = 22 = 0 H 0 : 02 0 , 12 0 , 22 0 Q=((SR-SU)/SU) * ([T-2(k+1)]/p) = ((8451-8393)/8393) * (170-2[3]/3)= 0.378 F 0.05 (3,164)= 2.66 Q < F Do not reject H 0 at = 0.05 level of significance. So there is no structural change.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
ECON 301 - Introduction to Econometrics I May 2013 METU - Department of Economics Instructor: Dr. Ozan ERUYGUR e-mail: [email protected] Lecture Notes 8 B. Predictive Test We have already discussed Chow’s test of structural stability in the previous lecture. In some cases the sub-period may be too small ( T 2 <k +1) and so Chow suggested the predictive or forecast test . Observation-Specific Dummy An observation-specific dummy is a dummy variable that takes on the value one for a specific observation and zero for all other observations. It is also called period-specific dummy (Kennedy, 2001, pp.226-227). (1) The coefficient estimate for the observation-specific dummy is the forecast error (prediction error) for that observation, and the estimated variance of this coefficient estimate is the estimate of the variance of the forecast error. (2) If the value of the dependent variable for the period (observation) in question is coded as zero instead of its actual value (which may not be known, if we are trying to forecast it) then the estimated coefficient of the period-specific dummy is the forecast of that period’s dependent variable. (3) By testing the estimated coefficient of the observation-specific dummy against zero, using t test, we can test whether or not that observation is “ consistent ” with the estimated relationship. An F
ECON 301 - Introduction to Econometrics I May 2013 METU - Department of Economics Instructor: Dr. Ozan ERUYGUR e-mail: [email protected] Lecture Notes 9 test (joint test) would be used to test if several observations could be considered consistent with the estimated equation. In this case each observation would have its own period-specific dummy. Such tests are sometimes called post-sample predictive tests. Consider the following model ( t =1991, 1992, …,2000): 0 1 1 2 2 2001 2001 2002 2002 2003 2003 t t t t Y X X e I e I e I u where the observation-specific dummies of 2001 I , 2002 I and 2003 I are defined as: 2001 1 for 2001 0 otherwise I , 2002 1 for 2002 0 otherwise I , 2003 1 for 2003 0 otherwise I Here, the coefficients of observation-specific dummies denoted by 2001 e , 2002 e and 2003 e are the expected value

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page7 / 19

METU Department of Economics Instructor Dr Ozan ERUYGUR e...

This preview shows document pages 7 - 10. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online