Can I call R function in Matlab?

Question

ABDULAZIZ ALTUN 2020-4-3

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/515173-can-i-call-r-function-in-matlab

评论： Hung Dao 2021-1-20

I am trying to minimize a function written in matlab using R function known by optim

The matlab function is as follow

%% Like_R: function description
function [outputs] = Like_R(x_lf,S_inter,r,mu,sigma_epsilon, x_init,varphi,C)
	% x_lf S the low frequency benchmark (i.e. annual stocks)
	% S_inter X the matrix of indicators
	% r the value for rho
	% mu the autocorraltion of the shock
	% sigma_epsilon sigma the standard error of the innovation
	% varphi gamma the vector of coefficients associated with the indicators in X
	c=size(S_inter,1);
	M_1 =r*diag(ones(c-1,1),-1);
	ey_M_1=inv(eye(c)-M_1);
	Omega=(mu.^abs(repmat((1:c),c,1)'-repmat((1:c),c,1)))*(sigma_epsilon^2)/(1-mu^2);
	Omega_x=ey_M_1*Omega*ey_M_1';
	M_2=zeros(c,1);
	M_2(1,1)=r;
	Theta=x_lf-C*ey_M_1*(M_2*x_init+S_inter*varphi);
	Log_likelihood = (-c/2)*log(2*pi) - c*log(sigma_epsilon^2)+ log(abs(1-mu^2)) - Theta'*inv(C*Omega_x*C')*Theta;
	outputs=-Log_likelihood;
end

I have entries for each of x_lf, S_inter, and C. However, I want to find the input elements that minimizes the above function. In R, if this function has been writen there one can simply define another function in R known by lev.uncons this:

lev.uncons<-function(y){
 Like_R(x_lf, S_inter,y[1],y[2],y[3], y[4],as.matrix(c(y[5],y[6])),C)
}

Then one can call the following function in R as

#you can use any optimisation algorithm
#with or without bondaries, priors, 
solDnc<-optim(c(0.9,0.9, 0.5,1000,0.3), lev.uncons, method =  "L-BFGS-B",
      lower = c(0.1,0, -0.5,0.1,0.01), upper = c(1,0.999,Inf,1), hessian = TRUE)   

What I want to do is to use matlab to call the minimization function optim in R. I know there is alternative such as min search or minimume constraint. But is there a way to do so from matlab?