Kalman filter starting value
Posted: Tue Sep 03, 2013 3:15 pm
Hi, everyone!
I am doing my thesis, and now meet a problem. When i run this model, I cannot get a good result. I guess there are something wrong with the starting value but I don't know how to fix it. Please help me. Thank you sooooo much!
@signal s = sv1*p + sv2*r + sv3*rpi + sv4*di + [var = exp(c(1))]
@state sv1 = sv1(-1) + [var = exp(c(2))]
@state sv2 = sv2(-1) + [var = exp(c(3))]
@state sv3 = sv3(-1) + [var = exp(c(4))]
@state sv4 = sv4(-1) + [var = exp(c(5))]
Sspace: UNTITLED
Method: Maximum likelihood (Marquardt)
Date: 09/04/13 Time: 00:04
Sample: 1998M04 2013M03
Included observations: 180
Convergence achieved after 1 iteration
WARNING: Singular covariance - coefficients are not unique
Coefficient Std. Error z-Statistic Prob.
C(1) -4.006168 NA NA NA
C(2) -9.980355 NA NA NA
C(3) -107.6388 NA NA NA
C(4) -51.74532 NA NA NA
C(5) -46.13349 NA NA NA
Final State Root MSE z-Statistic Prob.
SV1 2.432942 0.336807 7.223541 0.0000
SV2 -0.022845 0.377897 -0.060452 0.9518
SV3 -2.152555 0.632355 -3.404031 0.0007
SV4 0.263428 0.263323 1.000396 0.3171
Log likelihood 36.20017 Akaike info criterion -0.346669
Parameters 5 Schwarz criterion -0.257975
Diffuse priors 4 Hannan-Quinn criter. -0.310707
Here attached my data.
I am doing my thesis, and now meet a problem. When i run this model, I cannot get a good result. I guess there are something wrong with the starting value but I don't know how to fix it. Please help me. Thank you sooooo much!
@signal s = sv1*p + sv2*r + sv3*rpi + sv4*di + [var = exp(c(1))]
@state sv1 = sv1(-1) + [var = exp(c(2))]
@state sv2 = sv2(-1) + [var = exp(c(3))]
@state sv3 = sv3(-1) + [var = exp(c(4))]
@state sv4 = sv4(-1) + [var = exp(c(5))]
Sspace: UNTITLED
Method: Maximum likelihood (Marquardt)
Date: 09/04/13 Time: 00:04
Sample: 1998M04 2013M03
Included observations: 180
Convergence achieved after 1 iteration
WARNING: Singular covariance - coefficients are not unique
Coefficient Std. Error z-Statistic Prob.
C(1) -4.006168 NA NA NA
C(2) -9.980355 NA NA NA
C(3) -107.6388 NA NA NA
C(4) -51.74532 NA NA NA
C(5) -46.13349 NA NA NA
Final State Root MSE z-Statistic Prob.
SV1 2.432942 0.336807 7.223541 0.0000
SV2 -0.022845 0.377897 -0.060452 0.9518
SV3 -2.152555 0.632355 -3.404031 0.0007
SV4 0.263428 0.263323 1.000396 0.3171
Log likelihood 36.20017 Akaike info criterion -0.346669
Parameters 5 Schwarz criterion -0.257975
Diffuse priors 4 Hannan-Quinn criter. -0.310707
Here attached my data.