r - 如何从uGARCHfit(rugarch包)中提取AIC

Question

我使用 rugarch 包安装了 egarch 模型，并想从拟合模型中提取 AIC。我怎么做？

我尝试了两个代码fittedmodel@fit$infocriteria[1]和 fittedmodel@fit$criteria[1]但它们都不起作用

egarchspec=ugarchspec(variance.model = list(model = "eGARCH", garchOrder = c(1,1)),distribution.model="sged")
fittedmodel<-ugarchfit(egarchspec, data=pregfc$RAU)
fittedmodel@fit$infocriteria[1]

结果为NULL。

Answer 1

我们可以使用 infocriteria如

data(dmbp)
spec <- ugarchspec()
fit <- ugarchfit(data = dmbp[,1], spec = spec)

infocriteria(fit)
#                      
# Akaike       1.124508
# Bayes        1.141493
# Shibata      1.124490
# Hannan-Quinn 1.130749
infocriteria(fit)[1]
# [1] 1.124508

如果您希望更多地手动执行此操作或查看后面的公式，请参阅

getMethod("infocriteria", "uGARCHfit")

这导致

rugarch:::.information.test
# function (LLH, nObs, nPars) 
# {
#     AIC = (-2 * LLH)/nObs + 2 * nPars/nObs
#     BIC = (-2 * LLH)/nObs + nPars * log(nObs)/nObs
#     SIC = (-2 * LLH)/nObs + log((nObs + 2 * nPars)/nObs)
#     HQIC = (-2 * LLH)/nObs + (2 * nPars * log(log(nObs)))/nObs
#     informationTests = list(AIC = AIC, BIC = BIC, SIC = SIC, 
#         HQIC = HQIC)
#     return(informationTests)
# }
# <bytecode: 0x10e316fc0>
# <environment: namespace:rugarch>