前面的模型我已经做了修改,stoch_simul这部分得到的结果我已经基本满意了,
在做贝叶斯参数估计的时候,发生了这样的问题: (minus) the hessian matrix at the "mode" is not positive definite!
=> posterior variance of the estimated parameters are not positive.
You should try to change the initial values of the parameters using
the estimated_params_init block, or use another optimization routine.
Warning: The results below are most likely wrong!
以下是参数估计值
RESULTS FROM POSTERIOR MAXIMIZATION
parameters
prior mean mode s.d. t-stat prior pstdev
beta 0.990 0.9900 0.0020 492.2549 beta 0.0020
alpha 0.355 0.3550 0.0020 177.4950 beta 0.0020
delta 0.050 0.0500 0.0020 24.9791 beta 0.0020
rho_z 0.950 0.9500 0.0020 475.0000 norm 0.0020
psi 2.000 2.0000 0.0020 1000.0013 gamm 0.0020
rho_ms 0.970 0.9620 0.0002 4581.6554 beta 0.0020
rho_f 0.700 0.7126 0.0000 0.0000 beta 0.0020
rho_ex 0.700 0.7010 0.0000 0.0000 beta 0.0020
rho_g 0.700 0.7000 0.0020 349.9896 beta 0.0020
eta_f 2.000 1.9984 0.0008 2391.9386 norm 0.0500
eta_g 0.700 0.7057 0.0000 0.0000 norm 0.0020
k2 0.300 0.3000 0.0020 150.0000 norm 0.0020
h2 0.100 0.1000 0.0577 1.7330 gamm 0.0500
遇到这样的问题,我是不是只要更改那些t统计值不显著的参数,让他通过检验就行了?别的参数是否还需要修改。才能避过这类错误,让模型继续运行下去?


雷达卡





京公网安备 11010802022788号







