fixed_solution¶

Optimal Solution Returned by optimize_fixed¶

Syntax¶

CppAD::mixed::fixed_solution solution

Prototype¶

    struct fixed_solution {
        CppAD::vector<double>       fixed_opt;
        CppAD::vector<double>       fixed_lag;
        CppAD::vector<double>       fix_con_lag;
        CppAD::vector<double>       ran_con_lag;
        warm_start_struct           warm_start;
        CppAD::vector<trace_struct> trace_vec;
    };

Convention¶

If a Lagrange multiplier is non-zero (zero), the correspond constraint is active (is not active) at the optimal solution. The values specified below are as in the solution return by optimize_fixed .

fixed_opt¶

The size of this field is n_fixed . It is the final value (optimal value found) for the fixed effects.

fixed_lag¶

The size of this field is n_fixed . If solution . fixed_lag [ i ] is greater than zero (less than zero), it is the Lagrange multiplier for the upper (lower) bound for the i-th component of the fixed effects.

fix_con_lag¶

The size of this field is the number of fixed constraints; i.e., the size of the vector vec returned by the fix_constraint function. If solution . fix_con_lag [ i ] is greater than zero (less than zero), it is the Lagrange multiplier for the upper (lower) bound for the i-th component of the fixed constraint function.

ran_con_lag¶

The size of this field is the number of random constraints; i.e., the number of rows in the matrix A_rcv . If solution . ran_con_lag [ i ] is greater than zero (less than zero), it is the Lagrange multiplier for the upper (lower) bound for the i-th row of the random constraint matrix \(A\).

warm_start¶

This warm_start_struct contains the necessary information to continue the ipopt optimization from the current solution; i.e., warm start the optimization.

trace_vec¶

The i-th element of this vector is a trace_struct with the information corresponding to the i-th iteration of the optimization algorithm.