matlab dcc,求助:用matlab的dcc_mvgarch估计输出的结果不知如何解释

>> dcc_mvgarch(A,1,1,1,1)

Estimating GARCH model for Series 1

Warning: Options LargeScale = 'off' and Algorithm = 'trust-region-reflective' conflict.

Ignoring Algorithm and running active-set algorithm. To run trust-region-reflective, set

LargeScale = 'on'. To run active-set without this warning, use Algorithm = 'active-set'.

> In fmincon at 445

In fattailed_garch at 198

In dcc_mvgarch at 82

Estimating GARCH model for Series 2

Warning: Options LargeScale = 'off' and Algorithm = 'trust-region-reflective' conflict.

Ignoring Algorithm and running active-set algorithm. To run trust-region-reflective, set

LargeScale = 'on'. To run active-set without this warning, use Algorithm = 'active-set'.

> In fmincon at 445

In fattailed_garch at 198

In dcc_mvgarch at 82

Estimating the DCC model

Warning: Options LargeScale = 'off' and Algorithm = 'trust-region-reflective' conflict.

Ignoring Algorithm and running active-set algorithm. To run trust-region-reflective, set

LargeScale = 'on'. To run active-set without this warning, use Algorithm = 'active-set'.

> In fmincon at 445

In dcc_mvgarch at 98

____________________________________________________________

Diagnostic Information

Number of variables: 2

Functions

Objective:                            dcc_mvgarch_likelihood

Gradient:                             finite-differencing

Hessian:                              finite-differencing (or Quasi-Newton)

Constraints

Nonlinear constraints:             do not exist

Number of linear inequality constraints:    1

Number of linear equality constraints:      0

Number of lower bound constraints:          2

Number of upper bound constraints:          0

Algorithm selected

medium-scale: SQP, Quasi-Newton, line-search

____________________________________________________________

End diagnostic information

Max     Line search  Directional  First-order

Iter F-count        f(x)   constraint   steplength   derivative   optimality Procedure

0      3      371.557    -0.009998

1     11      370.344     -0.01937       0.0313         -140          312

2     21      370.257     -0.02603      0.00781        -34.8           82

3     25      370.082     -0.02081          0.5          -56         86.5

4     28      369.757     -0.02524            1        -52.8         6.83

5     31      369.751     -0.02493            1        -5.62        0.597

6     34      369.751     -0.02482            1       -0.645        0.151

7     37      369.751     -0.02481            1       -0.138       0.0341  Hessian modified

Local minimum possible. Constraints satisfied.

fmincon stopped because the size of the current search direction is less than

twice the default value of the step size tolerance and constraints were

satisfied to within the selected value of the constraint tolerance.

No active inequalities.

ans =

0.0001

0.1071

0.8498

0.0001

0.0429

0.9448

0.0297

0.9455

输入的数据为两时间序列,输入参数A为501×2的矩阵,上面为输出结果,ans=....那八个数字分别代表什么还搞不清楚,还有动态相关系数的数据该怎么得出,求高手解答?

  • 0
    点赞
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
DCC-MVGARCH (Dynamic Conditional Correlation - Multivariate GARCH) is a model used for estimating the conditional correlation matrix of a multivariate time series. It is an extension of the traditional GARCH model, which allows for time-varying correlation between the variables. In MATLAB, you can use the Econometrics Toolbox to estimate the DCC-MVGARCH model. The toolbox provides functions such as `dcc` and `mvnormpdf` that can be used for this purpose. Here is an example code snippet that shows how to estimate a DCC-MVGARCH model using MATLAB: ```matlab % Load the multivariate time series data load('data.mat'); % Specify the number of variables in the series numVariables = size(data, 2); % Specify the GARCH model specifications spec = garchset('R', 1, 'M', 1, 'Q', 1); % Estimate the conditional variances and residuals using univariate GARCH models [~, ~, ~, ~, residuals] = garchfit(spec, data); % Estimate the DCC-MVGARCH model specDCC = dccspec('VarianceModel', spec, 'Distribution', 't'); [estParams, ~, stdErrors, ~, LLF] = dccfit(residuals, specDCC); % Get the estimated conditional correlation matrix R = dccvar(estParams, residuals, specDCC); % Display the estimated parameters and standard errors disp(estParams); disp(stdErrors); % Display the log-likelihood value disp(LLF); % Display the conditional correlation matrix disp(R); ``` Please note that you need to have the Econometrics Toolbox installed in MATLAB to use these functions. Also, make sure to replace `'data.mat'` with the file name or data matrix containing your multivariate time series data. I hope this helps! Let me know if you have any further questions.

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值