经管之家首页 答疑首页 MATLAB 做bekk模型,怎样调用m文件,求助,挺急的T_T
  • 统计与数据分析
  • 经济学
  • 管理学
  • 金融学
  • 财会类
  • 国际贸易类
  • 考研考博与考证
  • 找数据和资料
  • 求职与职场
  • 学术与投稿类
  • 社会生活
  • 其它
大家好,目前我需要用matlab做garch-bekk模型,求两组金融高频数据的方差和协方差,以求套保比率。
现在,我已经找到了bekk的工具包,里面的程序如下,是一个m文件,我想知道,我怎样将我的两组数据带入,调用m文件,run出结果。
因为之前都没用过这个软件,想知道详细步骤,万分感谢,比较急,论坛币不多,希望能得到回复~或者用其它软件怎么做,有人知道么?

function [ll,lls,Ht] = bekk_likelihood(parameters,data,dataAsym,p,o,q,backCast,backCastAsym,type)
% Likelihood for BEKK(p,q) multivarate volatility model estimation
%
% USAGE:
%  [LL,LLS,HT] = bekk_likelihood(PARAMETERS,DATA,P,O,Q,BACKCAST,TYPE)
%
% INPUTS:
%   PARAMETERS   - Vector of parameters required to compute the (negative) of the log-likelihood
%   DATA         - K by K by T array of data
%   DATAASYM     - K by K by T array of asymmetric data
%   P            - Positive, scalar integer representing the number of symmetric innovations
%   O            - Non-negative, scalar integer representing the number of asymmetric innovations
%   Q            - Non-negative, scalar integer representing the number of conditional covariance lags
%   BACKCAST     - K by K matrix to use for back casting
%   TYPE         - Number indicating type:
%                    1 - Scalar
%                    2 - Diagonal
%                    3 - Full
%
% OUTPUTS:
%   LL           - The log likelihood evaluated at the PARAMETERS
%   LLS          - A T by 1 vector of log-likelihoods
%   HT           - A [K K T] dimension matrix of conditional covariances
%
% COMMENTS:
%
% See also BEKK

% Copyright: Kevin Sheppard
% kevin.sheppard@economics.ox.ac.uk
% Revision: 1    Date: 3/27/2012

% Get the parameters together
T = size(data,3);
k = size(data,2);

[C,A,G,B] = bekk_parameter_transform(parameters,p,o,q,k,type);

Ht = zeros(k,k,T);
lls = zeros(T,1);
logLikConst = k*log(2*pi);

for i=1:T
    Ht(:,:,i) = C;
    for j=1:p
        if (i-j)<=0
            Ht(:,:,i) = Ht(:,:,i) + A(:,:,j)'*backCast*A(:,:,j);
        else
            Ht(:,:,i) = Ht(:,:,i) + A(:,:,j)'*data(:,:,i-j)*A(:,:,j);
        end
    end
    for j=1:o
        if (i-j)<=0
            Ht(:,:,i) = Ht(:,:,i) + G(:,:,j)'*backCastAsym*G(:,:,j);
        else
            Ht(:,:,i) = Ht(:,:,i) + G(:,:,j)'*dataAsym(:,:,i-j)*G(:,:,j);
        end
    end   
    for j=1:q
        if (i-j)<=0
            Ht(:,:,i) = Ht(:,:,i) + B(:,:,j)'*backCast*B(:,:,j);
        else
            Ht(:,:,i) = Ht(:,:,i) + B(:,:,j)'*Ht(:,:,i-j)*B(:,:,j);
        end
    end
    lls(i) = 0.5*(logLikConst + log(det(Ht(:,:,i))) + sum(diag(Ht(:,:,i)^(-1)*data(:,:,i))));
end
ll = sum(lls);

if isnan(ll) || isinf(ll) || ~isreal(ll)
    ll = 1e7;
end

kimility
回答于 2015/03/30 15:21
matlab:
[NUM,TXT,RAM]=xls('data')
[parameters, loglikelihood, Ht,...]  = full_bekk_mvgarch(NUM,1,1)
 加载中...