Debian Codegraph @

tests/sandbox.mod

raw
1 // --+ options: json=compute, stochastic +--
2
3 @#ifndef simulate
4 @#define simulate = false
5 @#endif
6
7 var x1 x2 x1bar x2bar z y x u v s azertyuiopiop z1 z2 z3;
8
9 varexo ex1
10 ex2
11 ex1bar
12 ex2bar
13 ez
14 ey
15 ex
16 eu
17 ev
18 es;
19
20 parameters
21 rho_1 rho_2 rho_3 rho_4
22 a_x1_0 a_x1_1 a_x1_2 a_x1_x2_1 a_x1_x2_2
23 a_x2_0 a_x2_1 a_x2_2 a_x2_x1_1 a_x2_x1_2
24 e_c_m c_z_1 c_z_2 c_z_dx2 c_z_u c_z_dv c_z_s cx cy beta
25 lambda;
26
27 rho_1 = .9;
28 rho_2 = -.2;
29 rho_3 = .4;
30 rho_4 = -.3;
31
32
33 a_x1_0 = -.9;
34 a_x1_1 = .4;
35 a_x1_2 = .3;
36 a_x1_x2_1 = .1;
37 a_x1_x2_2 = .2;
38
39 a_x2_0 = -.9;
40 a_x2_1 = .2;
41 a_x2_2 = -.1;
42 a_x2_x1_1 = -.1;
43 a_x2_x1_2 = .2;
44
45 beta = .2;
46 e_c_m = .5;
47 c_z_1 = .2;
48 c_z_2 = -.1;
49 c_z_dx2 = .3;
50 c_z_u = .3;
51 c_z_dv = .4;
52 c_z_s = -.2;
53 cx = 1.0;
54 cy = 1.0;
55
56
57 lambda = 0.5; // Share of optimizing agents.
58
59 trend_component_model(model_name=toto, eqtags=['eq:x1', 'eq:x2', 'eq:x1bar', 'eq:x2bar'], targets=['eq:x1bar', 'eq:x2bar']);
60
61 pac_model(auxiliary_model_name=toto, discount=beta, model_name=pacman, auxname=rototo);
62
63 model;
64
65 [name='eq:u']
66 s = .3*s(-1) - .1*s(-2) + es;
67
68 [name='eq:diff(v)']
69 diff(v) = .5*diff(v(-1)) + ev;
70
71 [name='eq:u']
72 u = .5*u(-1) - .2*u(-2) + eu;
73
74 [name='eq:y']
75 y = rho_1*y(-1) + rho_2*y(-2) + ey;
76
77 [name='eq:x']
78 x = rho_3*x(-1) + rho_4*x(-2) + ex;
79
80 [name='eq:azertyuiopiop', rename='azertyuiopiop->qsdfghjklm']
81 azertyuiopiop = x + y;
82
83 [name='eq:x1']
84 diff(x1) = a_x1_0*(x1(-1)-x1bar(-1)) + a_x1_1*diff(x1(-1)) + a_x1_2*diff(x1(-2)) + a_x1_x2_1*diff(x2(-1)) + a_x1_x2_2*diff(x2(-2)) + ex1;
85
86 [name='eq:x2']
87 diff(x2) = a_x2_0*(x2(-1)-x2bar(-1)) + a_x2_1*diff(x1(-1)) + a_x2_2*diff(x1(-2)) + a_x2_x1_1*diff(x2(-1)) + a_x2_x1_2*diff(x2(-2)) + ex2;
88
89 [name='eq:x1bar']
90 x1bar = x1bar(-1) + ex1bar;
91
92 [name='eq:x2bar']
93 x2bar = x2bar(-1) + ex2bar;
94
95 [name='zpac']
96 diff(z) = lambda*(e_c_m*(x1(-1)-z(-1)) + c_z_1*diff(z(-1)) + c_z_2*diff(z(-2)) + pac_expectation(pacman) + c_z_s*s + c_z_dv*diff(v) ) + (1-lambda)*( cy*y + cx*x) + c_z_u*u + c_z_dx2*diff(x2) + ez;
97
98 [name='z1']
99 z1 = z+y-x+u;
100
101 [name='z2']
102 z2 = z-y+x-u;
103
104 [name='z3']
105 z3 = u-diff(v);
106
107 end;
108
109 shocks;
110 var ex1 = 1.0;
111 var ex2 = 1.0;
112 var ex1bar = 1.0;
113 var ex2bar = 1.0;
114 var ez = 1.0;
115 var ey = 0.1;
116 var ex = 0.1;
117 var eu = 0.05;
118 var ev = 0.05;
119 var es = 0.07;
120 end;
121
122 // Initialize the PAC model (build the Companion VAR representation for the auxiliary model).
123 pac.initialize('pacman');
124
125 // Update the parameters of the PAC expectation model (h0 and h1 vectors).
126 pac.update.expectation('pacman');
127
128 @#if simulate
129 verbatim;
130 % Simulate the model to create an artificial sample.
131 initialconditions = dseries(zeros(10, M_.endo_nbr+M_.exo_nbr), dates('2000Q1'), vertcat(M_.endo_names,M_.exo_names));
132 TrueData = simul_backward_model(initialconditions, 300);
133 ds = sandbox.dynamic_set_auxiliary_series(TrueData, M_.params);
134 TrueData.rototo = ds.rototo;
135 TrueData.save('simulateddata', 'm');
136 end;
137
138 @#endif