Because `analytic_derivation=1` by default, the Hessian is not computed in `dsge_likelihood`. `H0` being empty results in a crash >Error using * >Incorrect dimensions for matrix multiplication. Check that the number of columns in the first matrix matches the number of rows in the >second matrix. To operate on each element of the matrix individually, use TIMES (.*) for elementwise multiplication. > >Error in csminit1 (line 77) > dx = -H0*g; > ^ >Error in newrat (line 154) > [fval,x0] = csminit1(func0,xparam1,penalty,fval0(icount),gg,0,H,Verbose,varargin{:}); in the second iteration. The reason is that line 303 ``` [~,~,gg,hh]=penalty_objective_function(xparam1,func0,penalty,varargin{:}); ``` does not account for `hh` not being set (in contrast to previous checks like if `isempty(hh)`). The testsuite misses this due to only doing one iteration. [fs2000_analytic_derivation_5.mod](/uploads/816b79f9b07270eb4e4bb392e674b4c7/fs2000_analytic_derivation_5.mod)