Problems with Ramsey optimal policy

[Nick2382]

Hi all,

I have a somewhat adjusted model of Gerali et al. (2010) which runs using stoch_simul.
However, when trying to compute the Ramsey optimal policy in this model, I encounter some problems.

From earlier posts and examples I understand that I cannot use the steady; command before using ramsey_policy.
However, when using the initial values of Gerali and attempting to get the optimal policy, I get the following error (using Ramsey_Gerali.mod):

Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND =
6.114642e-33.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In lnsrch1 at 95
In solve1 at 107
In dynare_solve at 150
In dyn_ramsey_static at 66
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali at 1308
In dynare at 180
Warning: Matrix is singular to working precision.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In lnsrch1 at 95
In solve1 at 107
In dynare_solve at 150
In dyn_ramsey_static at 66
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali at 1308
In dynare at 180
Warning: Matrix is close to singular or badly scaled. Results may be inaccurate. RCOND =
9.176551e-38.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In lnsrch1 at 95
In solve1 at 107
In dynare_solve at 150
In dyn_ramsey_static at 66
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali at 1308
In dynare at 180
Reference to non-existent field 'ghx'.

Error in evaluate_planner_objective (line 46)
Gy = dr.ghx(nstatic+(1:nspred),:);

Error in ramsey_policy (line 40)
oo_.planner_objective_value = evaluate_planner_objective(M_,options_,oo_);

Error in Ramsey_Gerali (line 1308)
ramsey_policy(var_list_);

Error in dynare (line 180)
evalin('base',fname) ;

Could anyone explain what is happenning and how I can solve this problem?

Further, when I use the steady state values that are given as output as initial values, when I run stoch_simul of this model (using Ramsey_Gerali1), I get:

Warning: Rank deficient, rank = 0, tol = NaN.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In dynare_solve at 64
In dyn_ramsey_static at 66
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali1 at 1299
In dynare at 180
STEADY: numerical initial values or parameters incompatible with the following equations
Columns 1 through 15

1 2 4 5 6 7 10 11 12 13 16 17 18 19 20

Columns 16 through 23

28 29 30 31 32 39 46 67

Please check for example
i) if all parameters occurring in these equations are defined
ii) that no division by an endogenous variable initialized to 0 occurs
Warning: Rank deficient, rank = 0, tol = NaN.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In dyn_ramsey_static at 67
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali1 at 1299
In dynare at 180
Reference to non-existent field 'ys'.

Error in ramsey_policy (line 27)
oo_.steady_state = oo_.dr.ys;

Error in Ramsey_Gerali1 (line 1299)
ramsey_policy(var_list_);

Error in dynare (line 180)
evalin('base',fname) ;

Lastly, I tried to compute the steady state by hand to provide initial values for the steady (using Ramsey_Gerali_SS2), I get the following error message;

Warning: Matrix is singular, close to singular or badly scaled. Results may be inaccurate.
RCOND = NaN.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In dynare_solve at 64
In dyn_ramsey_static at 66
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali_SS2 at 1328
In dynare at 180
STEADY: numerical initial values or parameters incompatible with the following equations
2 7 11 32 33 34

Please check for example
i) if all parameters occurring in these equations are defined
ii) that no division by an endogenous variable initialized to 0 occurs
Warning: Matrix is singular to working precision.
> In dyn_ramsey_static>dyn_ramsey_static_1 at 152
In dyn_ramsey_static>@(x)dyn_ramsey_static_1(x,M,options_,oo) at 40
In dyn_ramsey_static at 67
In evaluate_steady_state at 55
In resol at 104
In stoch_simul at 88
In ramsey_policy at 25
In Ramsey_Gerali_SS2 at 1328
In dynare at 180
Reference to non-existent field 'ys'.

Error in ramsey_policy (line 27)
oo_.steady_state = oo_.dr.ys;

Error in Ramsey_Gerali_SS2 (line 1328)
ramsey_policy(var_list_);

Error in dynare (line 180)
evalin('base',fname) ;

My questions are as follow:
what approach is the best to use? providing numerical initial values or an analytical solution?
what do the errors mean and how can I solve these?

I am really looking forward to getting some help, thanks a lot in advance!!

Nick

[jpfeifer]

Dear Nick,
providing analytical values is usually the best way. Unfortunately, the diagnostic tools for Ramsey are still under development/buggy. I will try to look into this in the next days/weeks.