Garch 模型中的求解器失败

Question

晚上好！现在我用外部回归器测试了 GARCH 模型：

library("tseries")
library("xts")
library("rugarch")

Rim <- read.csv("Rim.csv",  header=T, sep=";")
Rcm <- read.csv("Rcm.csv",  header=T, sep=";")

n=119
spec <- ugarchspec(variance.model = list(model = "sGARCH", garchOrder = c(1, 1), submodel = NULL, external.regressors = NULL, variance.targeting = FALSE), mean.model = list(armaOrder = c(0, 0), external.regressors = matrix(Rcm[,1])), distribution.model = "norm", start.pars = list(), fixed.pars = list()) 
my_lms=lapply(1:n,function(x) ugarchfit(spec=spec,data=Rim[,x],solver.control=list(trace=0)))

昨天我测试了这个模型，R 显示了这个模型的结果。今天我没有更改代码，但现在 R 给出了错误：

Warning messages:
1: In .sgarchfit(spec = spec, data = data, out.sample = out.sample,  : 
ugarchfit-->warning: solver failer to converge.
2: In .sgarchfit(spec = spec, data = data, out.sample = out.sample,  : 
ugarchfit-->warning: solver failer to converge.

另外，我今天在出错后更新了这个软件包。我还说明了以下信息：

 sessionInfo()
R version 3.4.0 (2017-04-21)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)

Matrix products: default

locale:
[1] LC_COLLATE=Russian_Russia.1251  LC_CTYPE=Russian_Russia.1251    LC_MONETARY=Russian_Russia.1251
[4] LC_NUMERIC=C                    LC_TIME=Russian_Russia.1251    

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] rugarch_1.3-6 xts_0.9-7     zoo_1.8-0    

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.10                magrittr_1.5                knitr_1.15.1                misc3d_0.8-4               
 [5] xtable_1.8-2                lattice_0.20-35             R6_2.2.0                    FNN_1.1                    
 [9] Rsolnp_1.16                 GeneralizedHyperbolic_0.8-1 SkewHyperbolic_0.3-2        tools_3.4.0                
[13] spd_2.0-1                   grid_3.4.0                  KernSmooth_2.23-15          htmltools_0.3.6            
[17] digest_0.6.12               rgl_0.98.1                  numDeriv_2016.8-1           Matrix_1.2-9               
[21] shiny_1.0.3                 nloptr_1.0.4                DistributionUtils_0.5-1     ks_1.10.6                  
[25] htmlwidgets_0.8             codetools_0.2-15            mime_0.5                    compiler_3.4.0             
[29] multicool_0.1-10            expm_0.999-2                jsonlite_1.4                truncnorm_1.0-7            
[33] mvtnorm_1.0-6               httpuv_1.3.3    

请告诉我问题是什么，因为我无法理解这个错误。谢谢。

Answer 1

如果你将求解器设置为 ，hybrid它就会成功。看这个例子：

ugarchfit(data=file1[,3],spec=spec2,solver ='hybrid')