project里要用到广义误差分布的分位数 在坛里看到这问题讨论过 有牛人上传了编程 输出了结果 但是看不懂 请人看看 从下面的输出 哪项是分位数?或者怎么知道分位数?
LogL: LL_LL
Method: Maximum Likelihood (Marquardt)
Date: 10/20/09 Time: 20:26
Sample: 1 930
Included observations: 930
Evaluation order: By observation
Estimation settings: tol= 1.0e-05, derivs=accurate numeric
Initial Values: NU(1)=2.00000, MU(1)=0.00058, OMEGA(1)=2.7e-06,
ALPHA(1)=0.07885, BETA(1)=0.90363
Convergence achieved after 14 iterations
Coefficient Std. Error z-Statistic Prob.
NU(1) 1.258457 0.074498 16.89256 0.0000
MU(1) 0.000896 0.000293 3.054208 0.0023
OMEGA(1) 3.21E-06 1.50E-06 2.142452 0.0322
ALPHA(1) 0.070804 0.018888 3.748686 0.0002
BETA(1) 0.904666 0.023287 38.84901 0.0000
Log likelihood 2913.892 Akaike info criterion -6.255682
Avg. log likelihood 3.133217 Schwarz criterion -6.229687
Number of Coefs. 5 Hannan-Quinn criter. -6.245767
程序是:
Program to estimate a GARCH(1,1)-ged model on the series u.
'%path = @runpath+"../u/"
'cd %u
series d1 = 0
smpl @first @first
d1 = 1
smpl @all
'get starting values from Gaussian ARCH
equation eq1
eq1.arch u c
show eq1.output
'declare and innitialize parameters
coef(1) mu = eq1.c(1)
coef(1) omega = eq1.c(2)
coef(1) alpha =eq1.c(3)
coef(1) beta =eq1.c(4)
coef(1) nu=2
' Define pi
[email=!pi=@acos(-1]!pi=@acos(-1[/email])
'set up GARCH(1,1)-ged likelihood
logl ll_ll
ll_ll.append @logl logl
ll_ll.append [email=loggaml=@gammalog(1/nu(1]loggaml=@gammalog(1/nu(1[/email]))
ll_ll.append loglam=-log(2)/nu(1)+0.5*([email=loggaml-@gammalog(3/nu(1]loggaml-@gammalog(3/nu(1[/email])))
ll_ll.append res =u-mu(1)
ll_ll.append [email=sig2=@recode(d1=1,omega(1)/(1-alpha(1)-beta(1)),omega(1)+alpha(1)*res(-1)^2+beta(1)*sig2(-1]sig2=@recode(d1=1,omega(1)/(1-alpha(1)-beta(1)),omega(1)+alpha(1)*res(-1)^2+beta(1)*sig2(-1[/email]))
ll_ll.append [email=z=res/@sqrt(sig2]z=res/@sqrt(sig2[/email])
ll_ll.append logl =log(nu(1))-loglam-(1+1/nu(1))*log(2)-loggaml-0.5*@abs(z/exp(loglam))^nu(1)-0.5*log(sig2)
'estimate and display output
ll_ll.ml(showopts,m=1000,c=1e-5)
show ll_ll.output
scalar [email=ged=@qged(0.95,nu(1]ged=@qged(0.95,nu(1[/email]))
本文来自: 人大经济论坛 详细出处参考:http://www.pinggu.org/bbs/viewthread.php?tid=589256&page=1&from^^uid=1285898