[程序分享] Oxmetrics [推广有奖]

141楼

epoh 发表于 2012-6-3 10:59:02

zlhysb 发表于 2012-6-2 14:23
epoh老师,您好,非常感谢您为大家解答这么多，关于MS-var模型，有几个问题想请教您:1.请问我的OX6软件打开GN ...

问题一个一个来
问题一请参考19楼操作步骤

142楼

jinfang8866 发表于 2012-6-4 22:09:56

请教各位大侠，我用OX做的MGARCH一直出现下面的错误提示，找不到数据，是怎么回事？？？
--------------- Ox at 22:07:13 on 04-Jun-2012 ---------------

Ox Professional version 6.30 (Windows/U/MT) (C) J.A. Doornik, 1994-2011
Copyright for this package: S. Laurent, 2007-2010.
MG@RCH package version 1.3, object created on  4-06-2012
Copyright for this package: S. Laurent, 2000-2010.
G@RCH package version 6.1, object created on  4-06-2012
Database is not dated.
No database sample selected yet.
Warning: Failed to load data

OX code
#include <oxstd.h>
#include <oxdraw.h>
#import <packages/MGarch1/mgarch>

main()
{

//--- Ox code for MG@RCH( 1)
decl model = new MGarch();

model.Load("C:\\Program Files\\OxMetrics6.2\\Ox\\packages\\MGarch1\\data\\new02.in7");
model.Deterministic(-1);

model.Select(Y_VAR, {"r", 0, 0});
model.Select(Y_VAR, {"st", 0, 0});

model.CSTS(1,1);
model.DISTRI(NORMAL);
model.ARMA_ORDERS(1,0);
model.GARCH_ORDERS(1,1);
model.MODEL(CCC);
model.MLE(QMLE);
model.UGARCH_MODELS(GJR);
model.UGARCH_TRUNC(1000);
model.UGARCH_PrintOutput(1);
model.UGARCH_ARFIMA(0);
model.ONE_STEP(0);
model.TSE_LAGS(0);
model.SetSelSampleByDates(dayofcalendar(2003, 7), dayofcalendar(2008, 8));
model.Initialization(<>);
model.PrintOutput(1);
model.DoEstimation();

decl cov_vec,cor_vec,Varf,covf_vec,corf_vec;
model.INFO_CRITERIA(1);
model.NORMALITY_TEST(1);
model.Q_TEST(1,<5;10;20;50>);
model.Q2_TEST(1,<5;10;20;50>);
model.M_NORMALITY_TEST(1);
model.HOSKING_TEST(1,<5;10;20;50>);
model.HOSKING2_TEST(1,<5;10;20;50>);
model.LI_MCLEOD_TEST(1,<5;10;20;50>);
model.LI_MCLEOD2_TEST(1,<5;10;20;50>);
model.LMC_TEST(1);
model.ENGLE_SHEPPARD_TEST(1,<5;10>);
model.Tests();
model.PLOT_RAW(1);
model.PLOT_STAND_RES(1);
model.PLOT_VAR(1);
model.PLOT_COV(1);
model.PLOT_CORR(1);
model.Graphs_in_sample(-1);
model.FORECAST(1,10);
model.FORECASTING();

cov_vec=model.GetCov_vec();
for (decl i=0;i<columns(cov_vec);++i)
model.Append_in(cov_vec[][i],model.GetCovNames()[i]);
covf_vec=model.GetCovf_vec();
for (decl i=0;i<columns(covf_vec);++i)
model.Append_out(covf_vec[][i],model.GetCovfNames()[i]);
model.Save("C:\\Program Files\\OxMetrics6.2\\Ox\\packages\\MGarch1\\data\\new02.in7");

delete model;
}

143楼

jinfang8866 发表于 2012-6-4 22:14:38

我的数据是月度数据，是不是时间频率的问题？
望各位大侠指点迷津！！！

144楼

epoh 发表于 2012-6-5 19:48:24

jinfang8866 发表于 2012-6-4 22:09
请教各位大侠，我用OX做的MGARCH一直出现下面的错误提示，找不到数据，是怎么回事？？？
--------------- ...

你要先载入数据
file\open\...xxx.xls

145楼

jinfang8866 发表于 2012-6-9 19:38:13

epoh 发表于 2012-6-5 19:48
你要先载入数据
file\open\...xxx.xls

问题解决了，谢谢哈！

146楼

wjj0913 发表于 2012-6-23 15:58:11

好东西，谢谢分享！

147楼

haohao886 发表于 2012-7-16 04:36:06

内容出错，删除

148楼

haohao886 发表于 2012-7-16 04:39:05

epoh 发表于 2012-6-5 19:48
你要先载入数据
file\open\...xxx.xls

老师我用你给的EC3SLS模型分析了一段数据，但是在结果里出现了一个warning(WARNING in command 67 Procedure THSLS: At least one coefficient in the  table above could not be estimated due to singularity of the data or derivatives.)
而67列命令符为3SLS(inst=Xs) eqs;老师认为是哪个地方修改错了麽，还是因为数据本身的原因？中间修改过的部分为
const N,46 T,5;
set NT=N*T;

? Create the lists and matrices for the model to be estimated
? XYs, XY = all variables, including dependent, RHS, and instruments
? Xs,    = instruments (don't include C)
? eqs,  equations to be estimated (FRMLs; declare the PARAMs as usual)

? Fill in lists here
list XYs y1 y2 y3 z1 z2 x1 x2 x3;
list Xs z1 z2 x1 x2 x3;
list eqs eq1 eq2 eq3;

frml eq1 y1 = ay2*y2 + ay3*y3 + ax1*x1 + ax2*x2;
  param ay2 ay3 ax1 ax2;
frml eq2 y2 = by1*y1 + by3*y3 + bx3*x3 + bz1*z1;
  param by1 by3 bx3 bz1;
frml eq3 y3 = cy1*y1 + cy2*y2 + cz1*z1 + cz2*z2;
  param cy1 cy2 cz1 cz2;