Debian Codegraph @

tests/pac/trend-component-22/example1.mod

raw
1 // --+ options: json=compute, stochastic +--
2
3 var x1 x2 x1bar x2bar z y x;
4
5 varexo ex1 ex2 ex1bar ex2bar ez ey ex;
6
7 parameters
8 rho_1 rho_2 rho_3 rho_4
9 a_x1_0 a_x1_1 a_x1_2 a_x1_x2_1 a_x1_x2_2
10 a_x2_0 a_x2_1 a_x2_2 a_x2_x1_1 a_x2_x1_2
11 e_c_m c_z_1 c_z_2 c_z_dx2 beta
12 lambda;
13
14 rho_1 = .9;
15 rho_2 = -.2;
16 rho_3 = .4;
17 rho_4 = -.3;
18
19
20 a_x1_0 = -.9;
21 a_x1_1 = .4;
22 a_x1_2 = .3;
23 a_x1_x2_1 = .1;
24 a_x1_x2_2 = .2;
25
26 a_x2_0 = -.9;
27 a_x2_1 = .2;
28 a_x2_2 = -.1;
29 a_x2_x1_1 = -.1;
30 a_x2_x1_2 = .2;
31
32 beta = .2;
33 e_c_m = .5;
34 c_z_1 = .2;
35 c_z_2 = -.1;
36 c_z_dx2 = .3;
37
38 lambda = 0.5; // Share of optimizing agents.
39
40 trend_component_model(model_name=toto, eqtags=['eq:x1', 'eq:x2', 'eq:x1bar', 'eq:x2bar'], targets=['eq:x1bar', 'eq:x2bar']);
41
42 pac_model(auxiliary_model_name=toto, discount=beta, model_name=pacman);
43
44 model;
45
46 [name='eq:y']
47 y = rho_1*y(-1) + rho_2*y(-2) + ey;
48
49 [name='eq:x']
50 x = rho_3*x(-1) + rho_4*x(-2) + ex;
51
52 [name='eq:x1']
53 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;
54
55 [name='eq:x2']
56 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;
57
58 [name='eq:x1bar']
59 x1bar = x1bar(-1) + ex1bar;
60
61 [name='eq:x2bar']
62 x2bar = x2bar(-1) + ex2bar;
63
64 [name='zpac']
65 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)) + (1-lambda)*( y + x) + c_z_dx2*diff(x2) + ez;
66
67 end;
68
69 shocks;
70 var ex1 = 1.0;
71 var ex2 = 1.0;
72 var ex1bar = 1.0;
73 var ex2bar = 1.0;
74 var ez = 1.0;
75 var ey = 0.1;
76 var ex = 0.1;
77 end;
78
79 // Initialize the PAC model (build the Companion VAR representation for the auxiliary model).
80 pac.initialize('pacman');
81
82 // Update the parameters of the PAC expectation model (h0 and h1 vectors).
83 pac.update.expectation('pacman');
84
85 // Set initial conditions to zero. Please use more sensible values if any...
86 initialconditions = dseries(zeros(10, M_.endo_nbr+M_.exo_nbr), 2000Q1, vertcat(M_.endo_names,M_.exo_names));
87
88 // Simulate the model for 500 periods
89 TrueData = simul_backward_model(initialconditions, 5000);
90
91 // Define a structure describing the parameters to be estimated (with initial conditions).
92 clear eparams
93 eparams.e_c_m = .9;
94 eparams.c_z_1 = .5;
95 eparams.c_z_2 = .2;
96 eparams.c_z_dx2 = .5;
97
98 // Define the dataset used for estimation
99 edata = TrueData;
100 edata.ez = dseries(NaN(TrueData.nobs, 1), 2000Q1, 'ez');
101 pac.estimate.iterative_ols('zpac', eparams, edata, 2005Q1:2005Q1+4000);
102
103 e_c_m_iterative_ols = M_.params(strmatch('e_c_m', M_.param_names, 'exact'));
104 c_z_1_iterative_ols = M_.params(strmatch('c_z_1', M_.param_names, 'exact'));
105 c_z_2_iterative_ols = M_.params(strmatch('c_z_2', M_.param_names, 'exact'));
106 c_z_dx2_iterative_ols = M_.params(strmatch('c_z_dx2', M_.param_names, 'exact'));
107
108 disp(sprintf('Estimate of e_c_m: %f', e_c_m_iterative_ols))
109 disp(sprintf('Estimate of c_z_1: %f', c_z_1_iterative_ols))
110 disp(sprintf('Estimate of c_z_2: %f', c_z_2_iterative_ols))
111 disp(sprintf('Estimate of c_z_dx2: %f', c_z_dx2_iterative_ols))
112
113 skipline(2)
114
115 // Define a structure describing the parameters to be estimated (with initial conditions).
116 clear eparams
117 eparams.e_c_m = .9;
118 eparams.c_z_1 = .5;
119 eparams.c_z_2 = .2;
120 eparams.c_z_dx2 = .5;
121
122 edata = TrueData;
123 edata.ez = dseries(NaN(TrueData.nobs, 1), 2000Q1, 'ez');
124 pac.estimate.nls('zpac', eparams, edata, 2005Q1:2005Q1+4000, 'fmincon');
125
126 e_c_m_nls = M_.params(strmatch('e_c_m', M_.param_names, 'exact'));
127 c_z_1_nls = M_.params(strmatch('c_z_1', M_.param_names, 'exact'));
128 c_z_2_nls = M_.params(strmatch('c_z_2', M_.param_names, 'exact'));
129 c_z_dx2_nls = M_.params(strmatch('c_z_dx2', M_.param_names, 'exact'));
130
131 disp(sprintf('Estimate of e_c_m: %f', e_c_m_nls))
132 disp(sprintf('Estimate of c_z_1: %f', c_z_1_nls))
133 disp(sprintf('Estimate of c_z_2: %f', c_z_2_nls))
134 disp(sprintf('Estimate of c_z_dx2: %f', c_z_dx2_nls))
135
136 if abs(e_c_m_nls-e_c_m_iterative_ols)>.01
137 error('Iterative OLS and NLS do not provide consistent estimates (e_c_m)')
138 end
139
140 if abs(c_z_1_nls-c_z_1_iterative_ols)>.01
141 error('Iterative OLS and NLS do not provide consistent estimates (c_z_1)')
142 end
143
144 if abs(c_z_2_nls-c_z_2_iterative_ols)>.01
145 error('Iterative OLS and NLS do not provide consistent estimates (c_z_2)')
146 end
147
148 if abs(c_z_dx2_nls-c_z_dx2_iterative_ols)>.01
149 error('Iterative OLS and NLS do not provide consistent estimates (c_z_dx2)')
150 end