2022年GARCH模型在Matlab中的實(shí)現(xiàn)

1、多元 GARCH 模型推測(cè)的 Matlab 程序function parameters, loglikelihood, Ht, likelihoods, stdresid, stderrors, A, B, scores = full_bekk_mvgarchdata,p,q, BEKKoptions; % PURPOSE: % To Estimate a full BEKK multivariate GARCH model. % % % USAGE: % parameters, loglikelihood, Ht, likelihoods, stdresid, stderrors, A, B

2、, scores = full_bekk_mvgarchdata,p,q,options; % % % INPUTS: % data - A t by k matrix of zero mean residuals % p - The lag length of the innovation process % q - The lag length of the AR process % options - optional Options for the optimizationfminunc % % OUTPUTS: % parameters - A k*k+1/2+p*k2+q*k2 v

3、ector of estimated parameteters. F % or any k2 set of Innovation or AR parameters X, % reshapeX,k,k will give the correct matrix % To recover C, use ivechparmaeters1:k*k+1/2 % loglikelihood - The loglikelihood of the function at the optimum % Ht - A k x k x t 3 dimension matrix of conditional covari

4、ances % likelihoods - A t by 1 vector of individual likelihoods % stdresid - A t by k matrix of multivariate standardized residuals % stderrors - A numParams2 square matrix of robust Standad ErrorsA-1*B*A-1*t-1 % A - The estimated inverse of the non-robust Standard errors % B - The estimated covaria

5、nce of teh scores % scores - A t by numParams matrix of individual scores % need to try and get some smart startgin values if sizedata,2 sizedata,1 data=data; end t k=sizedata; k2=k*k+1/2; scalaropt=optimsetfminunc; scalaropt=optimsetscalaropt,TolFun,1e-1,Display,iter,Diagnostics,on,DiffMaxC hange,1

6、e-2; startingparameters=scalar_bekk_mvgarchdata,p,q,scalaropt; CChol=startingparameters1:k*k+1/2; C=ivechstartingparameters1:k*k+1/2*ivechstartingparameters1:k*k+1/2 ; newA=; newB=; for i=1:p newA=newA diagonesk,1*startingparametersk*k+1/2+i; end for i=1:q newB=newB diagonesk,1*startingparametersk*k

7、+1/2+i+p; end newA=reshapenewA,k*k*p,1; newB=reshapenewB,k*k*q,1; startingparameters=CChol;newA;newB; if nargin=6 A=hessian_2sidedfull_bekk_mvgarch_likelihood,parameters,data,p,q,k,k2,t; h=maxabsparameters/2,1e-2*eps1/3; hplus=parameters+h; hminus=parameters-h; likelihoodsplus=zerost,lengthparameter

8、s; likelihoodsminus=zerost,lengthparameters; for i=1:lengthparameters hparameters=parameters; hparametersi=hplusi; HOLDER, indivlike = full_bekk_mvgarch_likelihoodhparameters,data,p,q,k,k2,t; likelihoodsplus:,i=indivlike; end for i=1:lengthparameters hparameters=parameters; hparametersi=hminusi; HOLDER, indivlike = full_bekk_mvgarch_likelihoodhparameters,data,p,q,k,k2,t; likelihoo