7 | | mat V0 = 0.00001*eye(2); V0(0,0)= 0.1; // |
8 | | ARX Ar; Ar.set_statistics(1, V0, -1.0); |
9 | | |
10 | | mat mu(1,1); |
11 | | mat R(1,1); |
12 | | Ar._e()->mean_mat(mu,R); |
13 | | cout << "Prior moments: mu="<< mu << ", R=" << R <<endl; |
14 | | |
| 7 | mat V0 = 0.00001 * eye ( 2 ); |
| 8 | V0 ( 0, 0 ) = 0.1; // |
| 9 | ARX Ar; |
| 10 | Ar.set_statistics ( 1, V0, -1.0 ); |
| 11 | |
| 12 | mat mu ( 1, 1 ); |
| 13 | mat R ( 1, 1 ); |
| 14 | Ar._e()->mean_mat ( mu, R ); |
| 15 | cout << "Prior moments: mu=" << mu << ", R=" << R << endl; |
| 16 | |
23 | | cout << "Empirical moments: mu=" << sum(smp)/ndat << ", R=" << sum_sqr(smp)/ndat - pow(sum(smp)/ndat,2) << endl; |
24 | | Ar._e()->mean_mat(mu,R); |
25 | | cout << "Posterior moments: mu="<< mu << ", R=" << R <<endl; |
26 | | |
| 25 | cout << "Empirical moments: mu=" << sum ( smp ) / ndat << ", R=" << sum_sqr ( smp ) / ndat - pow ( sum ( smp ) / ndat, 2 ) << endl; |
| 26 | Ar._e()->mean_mat ( mu, R ); |
| 27 | cout << "Posterior moments: mu=" << mu << ", R=" << R << endl; |
| 28 | |
36 | | vec Ap_x=Ap->evallogcond_m(X,vec_1(1.0)); |
37 | | vec ll_x = Ar.logpred_m(X2); |
38 | | |
39 | | cout << "normalize : " << xstep*sum(exp(Ap_x)) << endl; |
40 | | cout << "normalize : " << xstep*sum(exp(ll_x)) << endl; |
41 | | |
42 | | it_file it("arx_elem_test.it"); |
43 | | it << Name("Ap_x") << Ap_x; |
44 | | it << Name("ll_x") << ll_x; |
| 38 | vec Ap_x = Ap->evallogcond_m ( X, vec_1 ( 1.0 ) ); |
| 39 | vec ll_x = Ar.logpred_m ( X2 ); |
| 40 | |
| 41 | cout << "normalize : " << xstep*sum ( exp ( Ap_x ) ) << endl; |
| 42 | cout << "normalize : " << xstep*sum ( exp ( ll_x ) ) << endl; |
| 43 | |
| 44 | it_file it ( "arx_elem_test.it" ); |
| 45 | it << Name ( "Ap_x" ) << Ap_x; |
| 46 | it << Name ( "ll_x" ) << ll_x; |