[程式] 關於RATs跑雙變量ANST-GARCH
[軟體程式類別]:
RATs
[程式問題]:
ANST-GARCH
[軟體熟悉度]:
中(3個月到1年)
[問題敘述]:
想請問有沒有人寫過雙變量ANST-GARCH的程式碼
因為最近在寫這個程式碼的時候遇到了一個奇怪的問題
就是跑一次程式碼跟跑第二次會跑出不同的結果
但是如果有關閉RATs程式之後，再重跑一次結果又會跟原本第一次的結果相同
我是以之前寫的雙變量GARCH-M模型當作基礎，
再加入smooth transition function的部分
[程式範例]:
compute n=2
compute gstart=1,gend=1555
*
dec series[vect] Y
dec frml[vect] R
dec vect[series] U(n)
dec series[vect] ECT
*
* The paths of the covariance matrices and uu' are saved in the
* SERIES[SYMM] named H and UU. UX and HX are used for the current values
* of the residual vector and H matrices
*
declare series[symm] G uu
*
* ux is used when extracting a u vector
*
declare symm Gx(n,n)
declare vect ux(n)
*
* These will be the parameters for the mean equations. These are adjusted to
add
* variance or covariance terms as needed.
*
dec vect EC(n) C1(n) C2(n) C3(n)
dec rect T1(n,n) T2(n,n) T3(n,n) F1(n,n) F2(n,n) F3(n,n)
dec frml[vect] Ts
dec frml[vect] Tf
dec frml[symm] Vs
dec frml[symm] Vf
*
*construct a error correct term
*
linreg LSP / ECTEQ
# constant LF1M
gset ECT = ||ECTEQ,ECTEQ||
nonlin(parmset=standmean) EC C1 T1 F1 C2 C3 T2 T3 F2 F3
nonlin(parmset=transmean) L
*
* Mean model = VAR(1) with sqrt(G) "M" term
*
frml R =
Y-EC.*ECT{1}-C1-C2.*Ts-C3.*Tf-T1*Y{1}-T2*Y{1}.*Ts-T3*Y{1}.*Tf-F1*%sqrt(%xdiag(G))-F2*%sqrt(%xdiag(G)).*Ts-F3*%sqrt(%xdiag(G)).*Tf
gset Y = ||dLSP|dLF1M||
*
* Run preliminary VAR(1) to get estimates of residuals
*
*
linreg dLSP / u(1)
# constant dLSP{1} dLF1M{1} ecteq{1}
linreg dLF1M / u(2)
# constant dLSP{1} dLF1M{1} ecteq{1}
*
linreg dLSP
# constant dLSP{1} DLF1M{1} u(1) u(2) ecteq{1}
compute C1(1)=%beta(1),$
T1(1,1)=%beta(2),$
T1(1,2)=%beta(3),$
F1(1,1)=%beta(4),$
F1(1,2)=%beta(5),$
EC(1)=%beta(6)
linreg dLF1M
# constant dLSP{1} dLF1M{1} u(1) u(2) ecteq{1}
compute C1(2)=%beta(1),$
T1(2,1)=%beta(2),$
T1(2,2)=%beta(3),$
F1(2,1)=%beta(4),$
F1(2,2)=%beta(5),$
EC(2)=%beta(6)
compute L = 14
vcv(matrix=WWW)
# u
*
* These are used to initialize pre-sample variances.
*
gset G * gend = WWW
gset uu * gend = WWW
set u(1) = 0.0
set u(2) = 0.0
*
declare frml[symm] Gf
*
frml logl = $
Gx = Gf(t) , $
G(t) = Gx, $
ux = R , $
uu(t) = %outerxx(ux), $
%pt(u,t,ux),$
%logdensity(Gx,ux)
*
*
*create smooth translation function
*
frml rs = u(1)
frml rf = u(2)
frml fs = %logistic(L*rs{1},1)
frml ff = %logistic(L*rf{1},1)
frml Ts = ||fs|fs||
frml Tf = ||ff|ff||
frml Vs = ||fs|fs,fs||
frml Vf = ||ff|ff,ff||
*
* This does a BEKK model for the variance
*
dec symm A1(n,n) B1(n,n) D1(n,n)
dec rect AA2(n,n) AA3(n,n) BB2(n,n) BB3(n,n) CC2(n,n) CC3(n,n)
compute AA2=.05*%identity(n),AA3=.50*%identity(n)
compute A1=%decomp(WWW)
nonlin(parmset=standgarch) A1 AA2 AA3 B1 BB2 BB3 D1 CC2 CC3
frml Gf =
A1*tr(A1)+AA3*G{1}*tr(AA3)+AA2*uu{1}*tr(AA2)+B1*tr(B1).*Vs+BB3*G{1}*tr(BB3).*Vs+BB2*uu{1}*tr(BB2).*Vs+D1*tr(D1).*Vf+CC3*G{1}*tr(CC3).*Vf+CC2*uu{1}*tr(CC2).*Vf
maximize(trace,parmset=standmean+standgarch,pmethod=simplex,piters=200,method=bfgs,iters=150)
logl gstart+2 gend
maximize(trace,parmset=standmean+transmean+standgarch,pmethod=simplex,piters=200,method=bfgs,iters=150)
logl gstart+2 gend
以上是很恐怖的程式碼 ><
-----------------------------------------------------------------------------
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 61.227.96.243
推
10/13 23:27, , 1F
10/13 23:27, 1F
→
10/16 20:57, , 2F
10/16 20:57, 2F
→
10/16 20:57, , 3F
10/16 20:57, 3F