# THÔNG TIN CODE #
# 1. CODE LỰA CHỌN BẬC ARIMA ĐƯỢC LẤY TỪ LUẬN ÁN TIẾN SĨ CỦA LÊ VĂN THỨ (2022) #
# 2. GIẢ ĐỊNH GARCH(1,1)-STD dựa theo https://ktpt.edu.vn/Uploads/Bai%20bao/2024/So%20320/1337.pdf #
library(rugarch)
## Warning: package 'rugarch' was built under R version 4.4.1
## Loading required package: parallel
##
## Attaching package: 'rugarch'
## The following object is masked from 'package:stats':
##
## sigma
library(fGarch)
## Warning: package 'fGarch' was built under R version 4.4.1
## NOTE: Packages 'fBasics', 'timeDate', and 'timeSeries' are no longer
## attached to the search() path when 'fGarch' is attached.
##
## If needed attach them yourself in your R script by e.g.,
## require("timeSeries")
library(readxl)
library(writexl)
## Warning: package 'writexl' was built under R version 4.4.1
library(ggplot2)
Data <- read_excel("C:/Users/use/Desktop/KLTN_LPTPT.xlsx", sheet = "RETURN")
Date <- Data$Date
ACB <- Data$ACB
BAB <- Data$BAB
BID <- Data$BID
CTG <- Data$CTG
EIB <- Data$EIB
HDB <- Data$HDB
MBB <- Data$MBB
MSB <- Data$MSB
NVB <- Data$NVB
OCB <- Data$OCB
SHB <- Data$SHB
SSB <- Data$SSB
STB <- Data$STB
TCB <- Data$TCB
TPB <- Data$TPB
VCB <- Data$VCB
VIB <- Data$VIB
VPB <- Data$VPB
# XÁC ĐỊNH BẬC AR(p) VÀ MA(q) BẰNG TIÊU CHÍ AIC #
print("ACB")
## [1] "ACB"
arima.ACB <- autoarfima(ACB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.ACB
## ma1 ma2 sigma
## 0.00000000 0.02719585 1.96232176
print("BAB")
## [1] "BAB"
arima.BAB <- autoarfima(BAB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.BAB
## mu ma1 ma2 sigma
## -0.07065558 -0.06272222 -0.09801884 1.27000898
print("BID")
## [1] "BID"
arima.BID <- autoarfima(BID, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.BID
## ar1 ar2 sigma
## -0.07825024 -0.07689418 2.14924481
print("CTG")
## [1] "CTG"
arima.CTG <- autoarfima(CTG, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.CTG
## ar1 ar2 ma1 ma2 sigma
## -1.3829863 -0.9727802 1.3667086 0.9376439 2.1351100
print("EIB")
## [1] "EIB"
arima.EIB <- autoarfima(EIB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.EIB
## ar1 ma1 sigma
## 0.7673825 -0.6830358 2.6020362
print("HDB")
## [1] "HDB"
arima.HDB <- autoarfima(HDB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.HDB
## ar1 ar2 ma1 ma2 sigma
## -0.06586148 0.78427715 0.00000000 -0.74845462 1.97262632
print("MBB")
## [1] "MBB"
arima.MBB <- autoarfima(MBB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.MBB
## ar1 ar2 ma1 ma2 sigma
## -0.2313770 -0.9321090 0.2346694 0.9736268 2.0752742
print("MSB")
## [1] "MSB"
arima.MSB <- autoarfima(MSB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.MSB
## ma1 sigma
## 0.1043534 2.0746136
print("NVB")
## [1] "NVB"
arima.NVB <- autoarfima(NVB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.NVB
## mu ma1 ma2 sigma
## -0.15442229 -0.08978626 -0.05452931 2.87546130
print("OCB")
## [1] "OCB"
arima.OCB <- autoarfima(OCB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.OCB
## ar1 ar2 ma1 ma2 sigma
## 1.2074828 -0.9776917 -1.2219076 1.0044619 2.6123903
print("SHB")
## [1] "SHB"
arima.SHB <- autoarfima(SHB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.SHB
## ar1 sigma
## 0.1011017 2.3572149
print("SSB")
## [1] "SSB"
arima.SSB <- autoarfima(SSB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.SSB
## mu ar1 ar2 ma1 ma2 sigma
## -0.2714749 0.0761409 0.9049090 -0.1456162 -0.8779182 1.7060577
print("STB")
## [1] "STB"
arima.STB <- autoarfima(STB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.STB
## ma1 ma2 sigma
## 0.00000000 -0.07291556 2.46695340
print("TCB")
## [1] "TCB"
arima.TCB <- autoarfima(TCB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.TCB
## ar1 sigma
## 0.06619723 3.31603219
print("TPB")
## [1] "TPB"
arima.TPB <- autoarfima(TPB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.TPB
## ma1 sigma
## -0.07090717 2.49323032
print("VCB")
## [1] "VCB"
arima.VCB <- autoarfima(VCB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.VCB
## ar1 ar2 ma1 ma2 sigma
## 0.0000000 -0.9536500 0.0000000 0.9317295 1.6181925
print("VIB")
## [1] "VIB"
arima.VIB <- autoarfima(VIB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.VIB
## mu sigma
## -0.08862467 2.22145505
print("VPB")
## [1] "VPB"
arima.VPB <- autoarfima(VPB, ar.max = 2, ma.max = 2, criterion = "AIC", method = "full")$fit@fit$coef
arima.VPB
## ar1 ar2 ma1 ma2 sigma
## 0.7280111 -0.8569683 -0.7677599 0.8000868 2.0178712
ACB.ar_order <- sum(!is.na(arima.ACB[c("ar1", "ar2")]))
ACB.ma_order <- sum(!is.na(arima.ACB[c("ma1", "ma2")]))
BAB.ar_order <- sum(!is.na(arima.BAB[c("ar1", "ar2")]))
BAB.ma_order <- sum(!is.na(arima.BAB[c("ma1", "ma2")]))
BID.ar_order <- sum(!is.na(arima.BID[c("ar1", "ar2")]))
BID.ma_order <- sum(!is.na(arima.BID[c("ma1", "ma2")]))
CTG.ar_order <- sum(!is.na(arima.CTG[c("ar1", "ar2")]))
CTG.ma_order <- sum(!is.na(arima.CTG[c("ma1", "ma2")]))
EIB.ar_order <- sum(!is.na(arima.EIB[c("ar1", "ar2")]))
EIB.ma_order <- sum(!is.na(arima.EIB[c("ma1", "ma2")]))
HDB.ar_order <- sum(!is.na(arima.HDB[c("ar1", "ar2")]))
HDB.ma_order <- sum(!is.na(arima.HDB[c("ma1", "ma2")]))
MBB.ar_order <- sum(!is.na(arima.MBB[c("ar1", "ar2")]))
MBB.ma_order <- sum(!is.na(arima.MBB[c("ma1", "ma2")]))
MSB.ar_order <- sum(!is.na(arima.MSB[c("ar1", "ar2")]))
MSB.ma_order <- sum(!is.na(arima.MSB[c("ma1", "ma2")]))
NVB.ar_order <- sum(!is.na(arima.NVB[c("ar1", "ar2")]))
NVB.ma_order <- sum(!is.na(arima.NVB[c("ma1", "ma2")]))
OCB.ar_order <- sum(!is.na(arima.OCB[c("ar1", "ar2")]))
OCB.ma_order <- sum(!is.na(arima.OCB[c("ma1", "ma2")]))
SHB.ar_order <- sum(!is.na(arima.SHB[c("ar1", "ar2")]))
SHB.ma_order <- sum(!is.na(arima.SHB[c("ma1", "ma2")]))
SSB.ar_order <- sum(!is.na(arima.SSB[c("ar1", "ar2")]))
SSB.ma_order <- sum(!is.na(arima.SSB[c("ma1", "ma2")]))
STB.ar_order <- sum(!is.na(arima.STB[c("ar1", "ar2")]))
STB.ma_order <- sum(!is.na(arima.STB[c("ma1", "ma2")]))
TCB.ar_order <- sum(!is.na(arima.TCB[c("ar1", "ar2")]))
TCB.ma_order <- sum(!is.na(arima.TCB[c("ma1", "ma2")]))
TPB.ar_order <- sum(!is.na(arima.TPB[c("ar1", "ar2")]))
TPB.ma_order <- sum(!is.na(arima.TPB[c("ma1", "ma2")]))
VCB.ar_order <- sum(!is.na(arima.VCB[c("ar1", "ar2")]))
VCB.ma_order <- sum(!is.na(arima.VCB[c("ma1", "ma2")]))
VIB.ar_order <- sum(!is.na(arima.VIB[c("ar1", "ar2")]))
VIB.ma_order <- sum(!is.na(arima.VIB[c("ma1", "ma2")]))
VPB.ar_order <- sum(!is.na(arima.VPB[c("ar1", "ar2")]))
VPB.ma_order <- sum(!is.na(arima.VPB[c("ma1", "ma2")]))
# MÔ HÌNH HÓA BIẾN ĐỘNG: GARCH, EGARCH, GJR-GARCH, APARCH #
print("ACB")
## [1] "ACB"
ACB.s <- ugarchspec(mean.model = list(armaOrder = c(ACB.ar_order,ACB.ma_order)),variance.model= list(garchOrder=c(1,1), model="sGARCH"), distribution.model = "std")
ACB.e <- ugarchspec(mean.model = list(armaOrder = c(ACB.ar_order,ACB.ma_order)),variance.model= list(garchOrder=c(1,1), model="eGARCH"), distribution.model = "std")
ACB.g <- ugarchspec(mean.model = list(armaOrder = c(ACB.ar_order,ACB.ma_order)),variance.model= list(garchOrder=c(1,1), model="gjrGARCH"), distribution.model = "std")
ACB.a <- ugarchspec(mean.model = list(armaOrder = c(ACB.ar_order,ACB.ma_order)),variance.model= list(garchOrder=c(1,1), model="apARCH"), distribution.model = "std")
ACB.s.11 <- ugarchfit(data=ACB, spec= ACB.s)
ACB.e.11 <- ugarchfit(data=ACB, spec= ACB.e)
ACB.g.11 <- ugarchfit(data=ACB, spec= ACB.g)
ACB.a.11 <- ugarchfit(data=ACB, spec= ACB.a)
sapply(list(ACB.s.11, ACB.e.11, ACB.g.11, ACB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 3.633537 3.629933 3.635187 3.626614
print("BAB")
## [1] "BAB"
BAB.s <- ugarchspec(mean.model = list(armaOrder = c(BAB.ar_order, BAB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
BAB.e <- ugarchspec(mean.model = list(armaOrder = c(BAB.ar_order, BAB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
BAB.g <- ugarchspec(mean.model = list(armaOrder = c(BAB.ar_order, BAB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
BAB.a <- ugarchspec(mean.model = list(armaOrder = c(BAB.ar_order, BAB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
BAB.s.11 <- ugarchfit(data = BAB, spec = BAB.s)
BAB.e.11 <- ugarchfit(data = BAB, spec = BAB.e)
BAB.g.11 <- ugarchfit(data = BAB, spec = BAB.g)
BAB.a.11 <- ugarchfit(data = BAB, spec = BAB.a)
sapply(list(BAB.s.11, BAB.e.11, BAB.g.11, BAB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 2.967204 2.978944 2.968303 2.970942
print("BID")
## [1] "BID"
BID.s <- ugarchspec(mean.model = list(armaOrder = c(BID.ar_order, BID.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
BID.e <- ugarchspec(mean.model = list(armaOrder = c(BID.ar_order, BID.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
BID.g <- ugarchspec(mean.model = list(armaOrder = c(BID.ar_order, BID.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
BID.a <- ugarchspec(mean.model = list(armaOrder = c(BID.ar_order, BID.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
BID.s.11 <- ugarchfit(data = BID, spec = BID.s)
BID.e.11 <- ugarchfit(data = BID, spec = BID.e)
BID.g.11 <- ugarchfit(data = BID, spec = BID.g)
BID.a.11 <- ugarchfit(data = BID, spec = BID.a)
sapply(list(BID.s.11, BID.e.11, BID.g.11, BID.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.131978 4.132072 4.131763 4.134142
print("CTG")
## [1] "CTG"
CTG.s <- ugarchspec(mean.model = list(armaOrder = c(CTG.ar_order, CTG.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
CTG.e <- ugarchspec(mean.model = list(armaOrder = c(CTG.ar_order, CTG.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
CTG.g <- ugarchspec(mean.model = list(armaOrder = c(CTG.ar_order, CTG.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
CTG.a <- ugarchspec(mean.model = list(armaOrder = c(CTG.ar_order, CTG.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
CTG.s.11 <- ugarchfit(data = CTG, spec = CTG.s)
CTG.e.11 <- ugarchfit(data = CTG, spec = CTG.e)
CTG.g.11 <- ugarchfit(data = CTG, spec = CTG.g)
CTG.a.11 <- ugarchfit(data = CTG, spec = CTG.a)
sapply(list(CTG.s.11, CTG.e.11, CTG.g.11, CTG.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.089352 4.081613 4.083732 4.082293
print("EIB")
## [1] "EIB"
EIB.s <- ugarchspec(mean.model = list(armaOrder = c(EIB.ar_order, EIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
EIB.e <- ugarchspec(mean.model = list(armaOrder = c(EIB.ar_order, EIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
EIB.g <- ugarchspec(mean.model = list(armaOrder = c(EIB.ar_order, EIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
EIB.a <- ugarchspec(mean.model = list(armaOrder = c(EIB.ar_order, EIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
EIB.s.11 <- ugarchfit(data = EIB, spec = EIB.s)
EIB.e.11 <- ugarchfit(data = EIB, spec = EIB.e)
EIB.g.11 <- ugarchfit(data = EIB, spec = EIB.g)
EIB.a.11 <- ugarchfit(data = EIB, spec = EIB.a)
sapply(list(EIB.s.11, EIB.e.11, EIB.g.11, EIB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.401624 4.407970 4.403810 4.405314
print("HDB")
## [1] "HDB"
HDB.s <- ugarchspec(mean.model = list(armaOrder = c(HDB.ar_order, HDB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
HDB.e <- ugarchspec(mean.model = list(armaOrder = c(HDB.ar_order, HDB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
HDB.g <- ugarchspec(mean.model = list(armaOrder = c(HDB.ar_order, HDB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
HDB.a <- ugarchspec(mean.model = list(armaOrder = c(HDB.ar_order, HDB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
HDB.s.11 <- ugarchfit(data = HDB, spec = HDB.s)
HDB.e.11 <- ugarchfit(data = HDB, spec = HDB.e)
HDB.g.11 <- ugarchfit(data = HDB, spec = HDB.g)
HDB.a.11 <- ugarchfit(data = HDB, spec = HDB.a)
sapply(list(HDB.s.11, HDB.e.11, HDB.g.11, HDB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 3.969168 3.952978 3.965488 3.966676
print("MBB")
## [1] "MBB"
MBB.s <- ugarchspec(mean.model = list(armaOrder = c(MBB.ar_order, MBB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
MBB.e <- ugarchspec(mean.model = list(armaOrder = c(MBB.ar_order, MBB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
MBB.g <- ugarchspec(mean.model = list(armaOrder = c(MBB.ar_order, MBB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
MBB.a <- ugarchspec(mean.model = list(armaOrder = c(MBB.ar_order, MBB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
MBB.s.11 <- ugarchfit(data = MBB, spec = MBB.s)
MBB.e.11 <- ugarchfit(data = MBB, spec = MBB.e)
MBB.g.11 <- ugarchfit(data = MBB, spec = MBB.g)
MBB.a.11 <- ugarchfit(data = MBB, spec = MBB.a)
sapply(list(MBB.s.11, MBB.e.11, MBB.g.11, MBB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 3.910550 3.900708 3.901392 3.909220
print("MSB")
## [1] "MSB"
MSB.s <- ugarchspec(mean.model = list(armaOrder = c(MSB.ar_order, MSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
MSB.e <- ugarchspec(mean.model = list(armaOrder = c(MSB.ar_order, MSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
MSB.g <- ugarchspec(mean.model = list(armaOrder = c(MSB.ar_order, MSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
MSB.a <- ugarchspec(mean.model = list(armaOrder = c(MSB.ar_order, MSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
MSB.s.11 <- ugarchfit(data = MSB, spec = MSB.s)
MSB.e.11 <- ugarchfit(data = MSB, spec = MSB.e)
MSB.g.11 <- ugarchfit(data = MSB, spec = MSB.g)
MSB.a.11 <- ugarchfit(data = MSB, spec = MSB.a)
sapply(list(MSB.s.11, MSB.e.11, MSB.g.11, MSB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 3.957554 3.948087 3.954681 3.954276
print("NVB")
## [1] "NVB"
NVB.s <- ugarchspec(mean.model = list(armaOrder = c(NVB.ar_order, NVB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
NVB.e <- ugarchspec(mean.model = list(armaOrder = c(NVB.ar_order, NVB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
NVB.g <- ugarchspec(mean.model = list(armaOrder = c(NVB.ar_order, NVB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
NVB.a <- ugarchspec(mean.model = list(armaOrder = c(NVB.ar_order, NVB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
NVB.s.11 <- ugarchfit(data = NVB, spec = NVB.s)
NVB.e.11 <- ugarchfit(data = NVB, spec = NVB.e)
NVB.g.11 <- ugarchfit(data = NVB, spec = NVB.g)
NVB.a.11 <- ugarchfit(data = NVB, spec = NVB.a)
sapply(list(NVB.s.11, NVB.e.11, NVB.g.11, NVB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.633119 4.628747 4.635963 4.637573
print("OCB")
## [1] "OCB"
OCB.s <- ugarchspec(mean.model = list(armaOrder = c(OCB.ar_order, OCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
OCB.e <- ugarchspec(mean.model = list(armaOrder = c(OCB.ar_order, OCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
OCB.g <- ugarchspec(mean.model = list(armaOrder = c(OCB.ar_order, OCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
OCB.a <- ugarchspec(mean.model = list(armaOrder = c(OCB.ar_order, OCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
OCB.s.11 <- ugarchfit(data = OCB, spec = OCB.s)
OCB.e.11 <- ugarchfit(data = OCB, spec = OCB.e)
OCB.g.11 <- ugarchfit(data = OCB, spec = OCB.g)
OCB.a.11 <- ugarchfit(data = OCB, spec = OCB.a)
sapply(list(OCB.s.11, OCB.e.11, OCB.g.11, OCB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.099220 4.088762 4.094254 4.068617
print("SHB")
## [1] "SHB"
SHB.s <- ugarchspec(mean.model = list(armaOrder = c(SHB.ar_order, SHB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
SHB.e <- ugarchspec(mean.model = list(armaOrder = c(SHB.ar_order, SHB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
SHB.g <- ugarchspec(mean.model = list(armaOrder = c(SHB.ar_order, SHB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
SHB.a <- ugarchspec(mean.model = list(armaOrder = c(SHB.ar_order, SHB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
SHB.s.11 <- ugarchfit(data = SHB, spec = SHB.s)
SHB.e.11 <- ugarchfit(data = SHB, spec = SHB.e)
SHB.g.11 <- ugarchfit(data = SHB, spec = SHB.g)
SHB.a.11 <- ugarchfit(data = SHB, spec = SHB.a)
sapply(list(SHB.s.11, SHB.e.11, SHB.g.11, SHB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.198206 4.185237 4.193084 4.190700
print("SSB")
## [1] "SSB"
SSB.s <- ugarchspec(mean.model = list(armaOrder = c(SSB.ar_order, SSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
SSB.e <- ugarchspec(mean.model = list(armaOrder = c(SSB.ar_order, SSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
SSB.g <- ugarchspec(mean.model = list(armaOrder = c(SSB.ar_order, SSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
SSB.a <- ugarchspec(mean.model = list(armaOrder = c(SSB.ar_order, SSB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
SSB.s.11 <- ugarchfit(data = SSB, spec = SSB.s)
SSB.e.11 <- ugarchfit(data = SSB, spec = SSB.e)
SSB.g.11 <- ugarchfit(data = SSB, spec = SSB.g)
SSB.a.11 <- ugarchfit(data = SSB, spec = SSB.a)
sapply(list(SSB.s.11, SSB.e.11, SSB.g.11, SSB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 2.905968 2.876373 2.899458 2.854939
print("STB")
## [1] "STB"
STB.s <- ugarchspec(mean.model = list(armaOrder = c(STB.ar_order, STB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
STB.e <- ugarchspec(mean.model = list(armaOrder = c(STB.ar_order, STB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
STB.g <- ugarchspec(mean.model = list(armaOrder = c(STB.ar_order, STB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
STB.a <- ugarchspec(mean.model = list(armaOrder = c(STB.ar_order, STB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
STB.s.11 <- ugarchfit(data = STB, spec = STB.s)
STB.e.11 <- ugarchfit(data = STB, spec = STB.e)
STB.g.11 <- ugarchfit(data = STB, spec = STB.g)
STB.a.11 <- ugarchfit(data = STB, spec = STB.a)
sapply(list(STB.s.11, STB.e.11, STB.g.11, STB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.408756 4.397059 4.399902 4.399614
print("TCB")
## [1] "TCB"
TCB.s <- ugarchspec(mean.model = list(armaOrder = c(TCB.ar_order, TCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
TCB.e <- ugarchspec(mean.model = list(armaOrder = c(TCB.ar_order, TCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
TCB.g <- ugarchspec(mean.model = list(armaOrder = c(TCB.ar_order, TCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
TCB.a <- ugarchspec(mean.model = list(armaOrder = c(TCB.ar_order, TCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
TCB.s.11 <- ugarchfit(data = TCB, spec = TCB.s)
TCB.e.11 <- ugarchfit(data = TCB, spec = TCB.e, solver = "lbfgs")
TCB.g.11 <- ugarchfit(data = TCB, spec = TCB.g)
TCB.a.11 <- ugarchfit(data = TCB, spec = TCB.a)
sapply(list(TCB.s.11, TCB.e.11,TCB.g.11, TCB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.170609 4.175192 4.173504 4.134520
print("TPB")
## [1] "TPB"
TPB.s <- ugarchspec(mean.model = list(armaOrder = c(TPB.ar_order, TPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
TPB.e <- ugarchspec(mean.model = list(armaOrder = c(TPB.ar_order, TPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
TPB.g <- ugarchspec(mean.model = list(armaOrder = c(TPB.ar_order, TPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
TPB.a <- ugarchspec(mean.model = list(armaOrder = c(TPB.ar_order, TPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
TPB.s.11 <- ugarchfit(data = TPB, spec = TPB.s)
TPB.e.11 <- ugarchfit(data = TPB, spec = TPB.e)
TPB.g.11 <- ugarchfit(data = TPB, spec = TPB.g)
TPB.a.11 <- ugarchfit(data = TPB, spec = TPB.a)
sapply(list(TPB.s.11, TPB.e.11, TPB.g.11, TPB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.144284 4.140224 4.147081 4.146757
print("VCB")
## [1] "VCB"
VCB.s <- ugarchspec(mean.model = list(armaOrder = c(VCB.ar_order, VCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
VCB.e <- ugarchspec(mean.model = list(armaOrder = c(VCB.ar_order, VCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
VCB.g <- ugarchspec(mean.model = list(armaOrder = c(VCB.ar_order, VCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
VCB.a <- ugarchspec(mean.model = list(armaOrder = c(VCB.ar_order, VCB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
VCB.s.11 <- ugarchfit(data = VCB, spec = VCB.s)
VCB.e.11 <- ugarchfit(data = VCB, spec = VCB.e)
VCB.g.11 <- ugarchfit(data = VCB, spec = VCB.g)
VCB.a.11 <- ugarchfit(data = VCB, spec = VCB.a)
sapply(list(VCB.s.11, VCB.e.11, VCB.g.11, VCB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 3.516838 3.534916 3.514585 3.512524
print("VIB")
## [1] "VIB"
VIB.s <- ugarchspec(mean.model = list(armaOrder = c(VIB.ar_order, VIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
VIB.e <- ugarchspec(mean.model = list(armaOrder = c(VIB.ar_order, VIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
VIB.g <- ugarchspec(mean.model = list(armaOrder = c(VIB.ar_order, VIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
VIB.a <- ugarchspec(mean.model = list(armaOrder = c(VIB.ar_order, VIB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
VIB.s.11 <- ugarchfit(data = VIB, spec = VIB.s)
VIB.e.11 <- ugarchfit(data = VIB, spec = VIB.e)
VIB.g.11 <- ugarchfit(data = VIB, spec = VIB.g)
VIB.a.11 <- ugarchfit(data = VIB, spec = VIB.a)
sapply(list(VIB.s.11, VIB.e.11, VIB.g.11, VIB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.061869 4.048364 4.064562 4.018601
print("VPB")
## [1] "VPB"
VPB.s <- ugarchspec(mean.model = list(armaOrder = c(VPB.ar_order, VPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "sGARCH"), distribution.model = "std")
VPB.e <- ugarchspec(mean.model = list(armaOrder = c(VPB.ar_order, VPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "eGARCH"), distribution.model = "std")
VPB.g <- ugarchspec(mean.model = list(armaOrder = c(VPB.ar_order, VPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "gjrGARCH"), distribution.model = "std")
VPB.a <- ugarchspec(mean.model = list(armaOrder = c(VPB.ar_order, VPB.ma_order)), variance.model = list(garchOrder = c(1, 1), model = "apARCH"), distribution.model = "std")
VPB.s.11 <- ugarchfit(data = VPB, spec = VPB.s)
VPB.e.11 <- ugarchfit(data = VPB, spec = VPB.e)
VPB.g.11 <- ugarchfit(data = VPB, spec = VPB.g)
VPB.a.11 <- ugarchfit(data = VPB, spec = VPB.a)
sapply(list(VPB.s.11, VPB.e.11, VPB.g.11, VPB.a.11), infocriteria)[-4,][-3,][-2,]
## [1] 4.047348 4.047418 4.049519 4.050252
# XÁC ĐỊNH MÔ HÌNH PHÙ HỢP NHẤT #
min_index <- which.min(sapply(list(ACB.s.11, ACB.e.11, ACB.g.11, ACB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("ACB.s.11", "ACB.e.11", "ACB.g.11", "ACB.a.11")
cat(models[min_index], "\n")
## ACB.a.11
min_index <- which.min(sapply(list(BAB.s.11, BAB.e.11, BAB.g.11, BAB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("BAB.s.11", "BAB.e.11", "BAB.g.11", "BAB.a.11")
cat(models[min_index], "\n")
## BAB.s.11
min_index <- which.min(sapply(list(BID.s.11, BID.e.11, BID.g.11, BID.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("BID.s.11", "BID.e.11", "BID.g.11", "BID.a.11")
cat(models[min_index], "\n")
## BID.g.11
min_index <- which.min(sapply(list(CTG.s.11, CTG.e.11, CTG.g.11, CTG.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("CTG.s.11", "CTG.e.11", "CTG.g.11", "CTG.a.11")
cat(models[min_index], "\n")
## CTG.e.11
min_index <- which.min(sapply(list(EIB.s.11, EIB.e.11, EIB.g.11, EIB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("EIB.s.11", "EIB.e.11", "EIB.g.11", "EIB.a.11")
cat(models[min_index], "\n")
## EIB.s.11
min_index <- which.min(sapply(list(HDB.s.11, HDB.e.11, HDB.g.11, HDB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("HDB.s.11", "HDB.e.11", "HDB.g.11", "HDB.a.11")
cat(models[min_index], "\n")
## HDB.e.11
min_index <- which.min(sapply(list(MBB.s.11, MBB.e.11, MBB.g.11, MBB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("MBB.s.11", "MBB.e.11", "MBB.g.11", "MBB.a.11")
cat(models[min_index], "\n")
## MBB.e.11
min_index <- which.min(sapply(list(MSB.s.11, MSB.e.11, MSB.g.11, MSB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("MSB.s.11", "MSB.e.11", "MSB.g.11", "MSB.a.11")
cat(models[min_index], "\n")
## MSB.e.11
min_index <- which.min(sapply(list(NVB.s.11, NVB.e.11, NVB.g.11, NVB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("NVB.s.11", "NVB.e.11", "NVB.g.11", "NVB.a.11")
cat(models[min_index], "\n")
## NVB.e.11
min_index <- which.min(sapply(list(OCB.s.11, OCB.e.11, OCB.g.11, OCB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("OCB.s.11", "OCB.e.11", "OCB.g.11", "OCB.a.11")
cat(models[min_index], "\n")
## OCB.a.11
min_index <- which.min(sapply(list(SHB.s.11, SHB.e.11, SHB.g.11, SHB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("SHB.s.11", "SHB.e.11", "SHB.g.11", "SHB.a.11")
cat(models[min_index], "\n")
## SHB.e.11
min_index <- which.min(sapply(list(SSB.s.11, SSB.e.11, SSB.g.11, SSB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("SSB.s.11", "SSB.e.11", "SSB.g.11", "SSB.a.11")
cat(models[min_index], "\n")
## SSB.a.11
min_index <- which.min(sapply(list(STB.s.11, STB.e.11, STB.g.11, STB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("STB.s.11", "STB.e.11", "STB.g.11", "STB.a.11")
cat(models[min_index], "\n")
## STB.e.11
min_index <- which.min(sapply(list(TCB.s.11, TCB.e.11, TCB.g.11, TCB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("TCB.s.11", "TCB.e.11", "TCB.g.11", "TCB.a.11")
cat(models[min_index], "\n")
## TCB.a.11
min_index <- which.min(sapply(list(TPB.s.11, TPB.e.11, TPB.g.11, TPB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("TPB.s.11", "TPB.e.11", "TPB.g.11", "TPB.a.11")
cat(models[min_index], "\n")
## TPB.e.11
min_index <- which.min(sapply(list(VCB.s.11, VCB.e.11, VCB.g.11, VCB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("VCB.s.11", "VCB.e.11", "VCB.g.11", "VCB.g.11")
cat(models[min_index], "\n")
## VCB.g.11
min_index <- which.min(sapply(list(VIB.s.11, VIB.e.11, VIB.g.11, VIB.a.11), infocriteria)[-4,][-3,][-2,])
models <- c("VIB.s.11", "VIB.e.11", "VIB.g.11", "VIB.a.11")
cat(models[min_index], "\n")
## VIB.a.11
min_index <- which.min(sapply(list(VPB.s.11, VPB.e.11, VPB.g.11, VPB.a.11),
infocriteria)[-4,][-3,][-2,])
models <- c("VPB.s.11", "VPB.e.11", "VPB.g.11", "VPB.a.11")
cat(models[min_index], "\n")
## VPB.s.11
# MÔ HÌNH PHÂN PHỐI BIÊN PHÙ HỢP NHẤT #
ACB.a.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : apARCH(1,1)
## Mean Model : ARFIMA(0,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043637 0.039742 1.09801 0.272198
## ma1 -0.013589 0.036982 -0.36745 0.713284
## ma2 -0.014083 0.033822 -0.41640 0.677118
## omega 0.092247 0.044110 2.09128 0.036503
## alpha1 0.201597 0.053954 3.73647 0.000187
## beta1 0.822162 0.045247 18.17060 0.000000
## gamma1 -0.043827 0.137810 -0.31803 0.750466
## delta 0.984946 0.270693 3.63861 0.000274
## shape 3.066992 0.400569 7.65659 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043637 0.039178 1.11381 0.265359
## ma1 -0.013589 0.040618 -0.33455 0.737963
## ma2 -0.014083 0.034307 -0.41051 0.681434
## omega 0.092247 0.048710 1.89379 0.058253
## alpha1 0.201597 0.057471 3.50783 0.000452
## beta1 0.822162 0.050438 16.30045 0.000000
## gamma1 -0.043827 0.175460 -0.24978 0.802754
## delta 0.984946 0.234004 4.20911 0.000026
## shape 3.066992 0.460433 6.66110 0.000000
##
## LogLikelihood : -1236.742
##
## Information Criteria
## ------------------------------------
##
## Akaike 3.6266
## Bayes 3.6860
## Shibata 3.6263
## Hannan-Quinn 3.6496
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.2083 0.6481
## Lag[2*(p+q)+(p+q)-1][5] 2.3359 0.8572
## Lag[4*(p+q)+(p+q)-1][9] 3.3997 0.8212
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.001625 0.9678
## Lag[2*(p+q)+(p+q)-1][5] 0.200220 0.9926
## Lag[4*(p+q)+(p+q)-1][9] 0.414749 0.9992
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.0001626 0.500 2.000 0.9898
## ARCH Lag[5] 0.0685213 1.440 1.667 0.9923
## ARCH Lag[7] 0.1695107 2.315 1.543 0.9980
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 2.3801
## Individual Statistics:
## mu 0.16752
## ma1 0.05114
## ma2 0.13392
## omega 0.19802
## alpha1 0.53738
## beta1 0.38697
## gamma1 0.45824
## delta 0.20780
## shape 0.73171
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.1 2.32 2.82
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 1.2243 0.2213
## Negative Sign Bias 0.2482 0.8041
## Positive Sign Bias 0.7680 0.4427
## Joint Effect 1.5512 0.6705
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 54.86 2.439e-05
## 2 30 100.90 7.008e-10
## 3 40 75.45 4.133e-04
## 4 50 103.17 9.839e-06
##
##
## Elapsed time : 2.315064
BAB.s.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : sGARCH(1,1)
## Mean Model : ARFIMA(0,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.067949 0.024993 -2.7187 0.006555
## ma1 -0.113370 0.040023 -2.8326 0.004617
## ma2 -0.120706 0.038966 -3.0977 0.001950
## omega 0.162890 0.061250 2.6594 0.007827
## alpha1 0.271863 0.080198 3.3899 0.000699
## beta1 0.672601 0.075487 8.9101 0.000000
## shape 3.809745 0.648895 5.8711 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.067949 0.025305 -2.6852 0.007248
## ma1 -0.113370 0.037860 -2.9944 0.002749
## ma2 -0.120706 0.032962 -3.6619 0.000250
## omega 0.162890 0.059604 2.7329 0.006278
## alpha1 0.271863 0.085988 3.1617 0.001569
## beta1 0.672601 0.079029 8.5108 0.000000
## shape 3.809745 0.637763 5.9736 0.000000
##
## LogLikelihood : -1012.234
##
## Information Criteria
## ------------------------------------
##
## Akaike 2.9672
## Bayes 3.0134
## Shibata 2.9670
## Hannan-Quinn 2.9851
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 1.140 0.285629
## Lag[2*(p+q)+(p+q)-1][5] 4.931 0.004183
## Lag[4*(p+q)+(p+q)-1][9] 7.001 0.127977
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.1967 0.6574
## Lag[2*(p+q)+(p+q)-1][5] 0.5368 0.9519
## Lag[4*(p+q)+(p+q)-1][9] 1.0449 0.9843
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.5501 0.500 2.000 0.4583
## ARCH Lag[5] 0.5635 1.440 1.667 0.8647
## ARCH Lag[7] 0.7502 2.315 1.543 0.9505
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 2.4171
## Individual Statistics:
## mu 0.47959
## ma1 0.44478
## ma2 0.02797
## omega 0.13414
## alpha1 0.35682
## beta1 0.26720
## shape 0.03995
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.8413 0.4005
## Negative Sign Bias 0.7247 0.4689
## Positive Sign Bias 0.8630 0.3884
## Joint Effect 1.2874 0.7321
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 125.0 1.267e-17
## 2 30 149.6 3.490e-18
## 3 40 152.2 2.717e-15
## 4 50 169.8 3.083e-15
##
##
## Elapsed time : 0.4976408
BID.g.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : gjrGARCH(1,1)
## Mean Model : ARFIMA(2,0,0)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.004869 0.052715 -0.09237 0.926404
## ar1 -0.120762 0.037127 -3.25267 0.001143
## ar2 -0.023387 0.034849 -0.67109 0.502166
## omega 0.042135 0.032607 1.29220 0.196289
## alpha1 0.039179 0.019751 1.98363 0.047297
## beta1 0.937942 0.019178 48.90797 0.000000
## gamma1 0.035665 0.026106 1.36617 0.171885
## shape 3.935382 0.638110 6.16725 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.004869 0.053908 -0.090325 0.928029
## ar1 -0.120762 0.038271 -3.155401 0.001603
## ar2 -0.023387 0.033056 -0.707488 0.479263
## omega 0.042135 0.034150 1.233829 0.217267
## alpha1 0.039179 0.018080 2.167020 0.030233
## beta1 0.937942 0.020212 46.404273 0.000000
## gamma1 0.035665 0.029925 1.191791 0.233343
## shape 3.935382 0.583452 6.744996 0.000000
##
## LogLikelihood : -1411.261
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.1318
## Bayes 4.1845
## Shibata 4.1315
## Hannan-Quinn 4.1522
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 2.592 0.10743
## Lag[2*(p+q)+(p+q)-1][5] 4.000 0.06652
## Lag[4*(p+q)+(p+q)-1][9] 6.221 0.22222
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.0003576 0.9849
## Lag[2*(p+q)+(p+q)-1][5] 3.1897357 0.3735
## Lag[4*(p+q)+(p+q)-1][9] 5.4380917 0.3680
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 1.436 0.500 2.000 0.2307
## ARCH Lag[5] 4.756 1.440 1.667 0.1170
## ARCH Lag[7] 5.576 2.315 1.543 0.1724
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.0049
## Individual Statistics:
## mu 0.04392
## ar1 0.07741
## ar2 0.37992
## omega 0.10690
## alpha1 0.13773
## beta1 0.13494
## gamma1 0.22044
## shape 0.09779
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.89 2.11 2.59
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.9803 0.3273
## Negative Sign Bias 0.6163 0.5379
## Positive Sign Bias 0.3232 0.7466
## Joint Effect 1.0354 0.7927
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 23.25 0.2266
## 2 30 20.29 0.8836
## 3 40 32.83 0.7464
## 4 50 38.25 0.8661
##
##
## Elapsed time : 1.092271
CTG.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.041162 0.055040 0.74787 0.45454
## ar1 -1.396818 0.001510 -925.27702 0.00000
## ar2 -0.996216 0.004325 -230.34500 0.00000
## ma1 1.392576 0.003358 414.70882 0.00000
## ma2 0.993043 0.001787 555.56822 0.00000
## omega 0.040940 0.002618 15.63570 0.00000
## alpha1 -0.071759 0.031031 -2.31251 0.02075
## beta1 0.968422 0.002380 406.95278 0.00000
## gamma1 0.223948 0.042285 5.29614 0.00000
## shape 4.093053 0.536719 7.62607 0.00000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.041162 0.051130 0.80506 0.420785
## ar1 -1.396818 0.001390 -1004.92963 0.000000
## ar2 -0.996216 0.012813 -77.74834 0.000000
## ma1 1.392576 0.010292 135.30599 0.000000
## ma2 0.993043 0.005971 166.31616 0.000000
## omega 0.040940 0.008000 5.11737 0.000000
## alpha1 -0.071759 0.029955 -2.39553 0.016596
## beta1 0.968422 0.002326 416.26057 0.000000
## gamma1 0.223948 0.075113 2.98150 0.002868
## shape 4.093053 1.169434 3.50003 0.000465
##
## LogLikelihood : -1392.034
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.0816
## Bayes 4.1476
## Shibata 4.0812
## Hannan-Quinn 4.1071
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.8168 0.3661
## Lag[2*(p+q)+(p+q)-1][11] 3.2546 1.0000
## Lag[4*(p+q)+(p+q)-1][19] 7.0868 0.9005
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.1195 0.7296
## Lag[2*(p+q)+(p+q)-1][5] 0.7051 0.9219
## Lag[4*(p+q)+(p+q)-1][9] 1.8783 0.9190
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.1381 0.500 2.000 0.7102
## ARCH Lag[5] 0.8110 1.440 1.667 0.7897
## ARCH Lag[7] 1.1659 2.315 1.543 0.8852
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.3814
## Individual Statistics:
## mu 0.16183
## ar1 0.04856
## ar2 0.03908
## ma1 0.05375
## ma2 0.05469
## omega 0.13557
## alpha1 0.32710
## beta1 0.29239
## gamma1 0.03212
## shape 0.38672
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.29 2.54 3.05
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 1.2213 0.2224
## Negative Sign Bias 0.2613 0.7939
## Positive Sign Bias 0.7338 0.4633
## Joint Effect 1.6311 0.6524
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 12.01 0.88519
## 2 30 27.28 0.55663
## 3 40 56.00 0.03811
## 4 50 56.74 0.20875
##
##
## Elapsed time : 2.161631
EIB.s.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : sGARCH(1,1)
## Mean Model : ARFIMA(1,0,1)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.032171 0.043354 -0.74206 0.458053
## ar1 0.925656 0.020616 44.89980 0.000000
## ma1 -0.956962 0.011315 -84.57447 0.000000
## omega 0.240714 0.163801 1.46955 0.141685
## alpha1 0.200505 0.062749 3.19533 0.001397
## beta1 0.798495 0.064611 12.35848 0.000000
## shape 3.615454 0.468917 7.71022 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.032171 0.048926 -0.65754 0.510831
## ar1 0.925656 0.019382 47.75975 0.000000
## ma1 -0.956962 0.008346 -114.66435 0.000000
## omega 0.240714 0.274884 0.87569 0.381197
## alpha1 0.200505 0.082075 2.44296 0.014567
## beta1 0.798495 0.100822 7.91988 0.000000
## shape 3.615454 0.451586 8.00613 0.000000
##
## LogLikelihood : -1504.958
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.4016
## Bayes 4.4478
## Shibata 4.4014
## Hannan-Quinn 4.4195
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 1.497 0.2211
## Lag[2*(p+q)+(p+q)-1][5] 3.441 0.2310
## Lag[4*(p+q)+(p+q)-1][9] 4.932 0.4690
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.02294 0.8796
## Lag[2*(p+q)+(p+q)-1][5] 0.41785 0.9696
## Lag[4*(p+q)+(p+q)-1][9] 1.15458 0.9789
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.05698 0.500 2.000 0.8113
## ARCH Lag[5] 0.18543 1.440 1.667 0.9689
## ARCH Lag[7] 0.73857 2.315 1.543 0.9520
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.95
## Individual Statistics:
## mu 0.14497
## ar1 0.25747
## ma1 0.24093
## omega 0.06803
## alpha1 0.59714
## beta1 0.20413
## shape 0.24231
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.4065 0.6845
## Negative Sign Bias 0.7642 0.4450
## Positive Sign Bias 0.6587 0.5103
## Joint Effect 1.3192 0.7246
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 19.81 0.40596
## 2 30 42.91 0.04638
## 3 40 49.59 0.11911
## 4 50 58.78 0.15983
##
##
## Elapsed time : 0.469908
HDB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.316270 0.000130 -2431.1682 0.0e+00
## ar1 1.146939 0.000149 7718.4457 0.0e+00
## ar2 -0.148121 0.000048 -3057.7726 0.0e+00
## ma1 -1.239952 0.000104 -11967.5138 0.0e+00
## ma2 0.231315 0.000035 6689.5140 0.0e+00
## omega 0.027565 0.002627 10.4942 0.0e+00
## alpha1 -0.047932 0.011802 -4.0613 4.9e-05
## beta1 0.982205 0.000445 2206.7213 0.0e+00
## gamma1 0.125852 0.007698 16.3492 0.0e+00
## shape 2.898225 0.164004 17.6716 0.0e+00
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.316270 0.001118 -283.00310 0.00000
## ar1 1.146939 0.003530 324.87909 0.00000
## ar2 -0.148121 0.000530 -279.36037 0.00000
## ma1 -1.239952 0.002608 -475.39299 0.00000
## ma2 0.231315 0.001312 176.33948 0.00000
## omega 0.027565 0.161021 0.17119 0.86408
## alpha1 -0.047932 0.460432 -0.10410 0.91709
## beta1 0.982205 0.008500 115.55135 0.00000
## gamma1 0.125852 0.652353 0.19292 0.84702
## shape 2.898225 7.535529 0.38461 0.70053
##
## LogLikelihood : -1347.848
##
## Information Criteria
## ------------------------------------
##
## Akaike 3.9530
## Bayes 4.0190
## Shibata 3.9526
## Hannan-Quinn 3.9785
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.02173 8.828e-01
## Lag[2*(p+q)+(p+q)-1][11] 9.60900 1.557e-07
## Lag[4*(p+q)+(p+q)-1][19] 16.69318 7.644e-03
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.0009577 0.9753
## Lag[2*(p+q)+(p+q)-1][5] 0.2853987 0.9852
## Lag[4*(p+q)+(p+q)-1][9] 0.7194397 0.9950
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.1658 0.500 2.000 0.6839
## ARCH Lag[5] 0.5285 1.440 1.667 0.8752
## ARCH Lag[7] 0.6526 2.315 1.543 0.9626
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.858
## Individual Statistics:
## mu 0.04979
## ar1 0.04946
## ar2 0.05036
## ma1 0.04413
## ma2 0.04915
## omega 0.18523
## alpha1 0.08505
## beta1 0.21061
## gamma1 0.13543
## shape 0.36521
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.29 2.54 3.05
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 1.3673 0.1720
## Negative Sign Bias 0.8975 0.3698
## Positive Sign Bias 1.0727 0.2838
## Joint Effect 2.2749 0.5173
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 28.20 0.07972
## 2 30 45.01 0.02937
## 3 40 50.29 0.10629
## 4 50 58.34 0.16953
##
##
## Elapsed time : 2.646098
MBB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.087275 0.051502 1.6946 0.090148
## ar1 -0.217366 0.014153 -15.3585 0.000000
## ar2 -0.954804 0.009620 -99.2511 0.000000
## ma1 0.226123 0.008402 26.9143 0.000000
## ma2 0.980963 0.001474 665.4408 0.000000
## omega 0.058414 0.032303 1.8083 0.070554
## alpha1 -0.116607 0.042312 -2.7559 0.005854
## beta1 0.952957 0.021550 44.2215 0.000000
## gamma1 0.238618 0.062053 3.8454 0.000120
## shape 3.413996 0.496676 6.8737 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.087275 0.051219 1.7039 0.088391
## ar1 -0.217366 0.013192 -16.4769 0.000000
## ar2 -0.954804 0.009227 -103.4740 0.000000
## ma1 0.226123 0.007494 30.1751 0.000000
## ma2 0.980963 0.001171 837.3731 0.000000
## omega 0.058414 0.038571 1.5145 0.129907
## alpha1 -0.116607 0.046333 -2.5167 0.011846
## beta1 0.952957 0.027394 34.7876 0.000000
## gamma1 0.238618 0.052566 4.5394 0.000006
## shape 3.413996 0.409840 8.3301 0.000000
##
## LogLikelihood : -1329.893
##
## Information Criteria
## ------------------------------------
##
## Akaike 3.9007
## Bayes 3.9667
## Shibata 3.9003
## Hannan-Quinn 3.9262
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.8873 0.3462
## Lag[2*(p+q)+(p+q)-1][11] 4.2518 0.9993
## Lag[4*(p+q)+(p+q)-1][19] 6.1438 0.9647
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.2509 0.6164
## Lag[2*(p+q)+(p+q)-1][5] 1.4970 0.7406
## Lag[4*(p+q)+(p+q)-1][9] 3.0763 0.7463
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.0690 0.500 2.000 0.7928
## ARCH Lag[5] 0.7087 1.440 1.667 0.8208
## ARCH Lag[7] 1.9572 2.315 1.543 0.7264
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.546
## Individual Statistics:
## mu 0.03106
## ar1 0.47419
## ar2 0.01670
## ma1 0.36999
## ma2 0.02714
## omega 0.11790
## alpha1 0.42278
## beta1 0.18885
## gamma1 0.05204
## shape 0.04647
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.29 2.54 3.05
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.5928 0.5535
## Negative Sign Bias 0.5630 0.5736
## Positive Sign Bias 1.0373 0.3000
## Joint Effect 1.4588 0.6918
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 18.71 0.4758
## 2 30 30.42 0.3931
## 3 40 31.66 0.7919
## 4 50 41.75 0.7592
##
##
## Elapsed time : 1.177285
MSB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(0,0,1)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.060553 0.052547 1.15236 0.249175
## ma1 0.032700 0.032980 0.99151 0.321437
## omega 0.237900 0.121868 1.95211 0.050925
## alpha1 -0.183842 0.083665 -2.19736 0.027995
## beta1 0.857550 0.060847 14.09360 0.000000
## gamma1 0.345961 0.118340 2.92345 0.003462
## shape 2.641884 0.341911 7.72682 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.060553 0.054365 1.1138 0.265350
## ma1 0.032700 0.029925 1.0927 0.274511
## omega 0.237900 0.218521 1.0887 0.276295
## alpha1 -0.183842 0.119427 -1.5394 0.123717
## beta1 0.857550 0.111827 7.6686 0.000000
## gamma1 0.345961 0.165840 2.0861 0.036968
## shape 2.641884 0.373531 7.0727 0.000000
##
## LogLikelihood : -1349.168
##
## Information Criteria
## ------------------------------------
##
## Akaike 3.9481
## Bayes 3.9943
## Shibata 3.9479
## Hannan-Quinn 3.9660
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 1.999 0.15737
## Lag[2*(p+q)+(p+q)-1][2] 2.493 0.09205
## Lag[4*(p+q)+(p+q)-1][5] 2.972 0.43826
## d.o.f=1
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 4.139 0.04191
## Lag[2*(p+q)+(p+q)-1][5] 4.859 0.16469
## Lag[4*(p+q)+(p+q)-1][9] 5.822 0.31913
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.07749 0.500 2.000 0.7807
## ARCH Lag[5] 1.22317 1.440 1.667 0.6680
## ARCH Lag[7] 1.58743 2.315 1.543 0.8037
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 2.5614
## Individual Statistics:
## mu 0.05181
## ma1 0.59676
## omega 0.16494
## alpha1 0.41335
## beta1 0.40106
## gamma1 0.05790
## shape 0.26067
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.01527 0.9878
## Negative Sign Bias 0.27335 0.7847
## Positive Sign Bias 0.83388 0.4046
## Joint Effect 1.29610 0.7301
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 172.9 7.539e-27
## 2 30 194.6 1.907e-26
## 3 40 193.3 2.520e-22
## 4 50 253.1 2.696e-29
##
##
## Elapsed time : 0.7386749
NVB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(0,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.20576 0.051870 -3.966797 0.000073
## ma1 -0.12582 0.040752 -3.087462 0.002019
## ma2 -0.07935 0.033061 -2.400139 0.016389
## omega 0.52238 0.190966 2.735459 0.006229
## alpha1 -0.00270 0.077589 -0.034805 0.972235
## beta1 0.79135 0.064738 12.223753 0.000000
## gamma1 0.80953 0.198156 4.085301 0.000044
## shape 2.55045 0.308890 8.256830 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.20576 0.051633 -3.985014 0.000067
## ma1 -0.12582 0.041942 -2.999877 0.002701
## ma2 -0.07935 0.026641 -2.978512 0.002897
## omega 0.52238 0.215252 2.426825 0.015232
## alpha1 -0.00270 0.070287 -0.038421 0.969352
## beta1 0.79135 0.074436 10.631297 0.000000
## gamma1 0.80953 0.177844 4.551891 0.000005
## shape 2.55045 0.262319 9.722718 0.000000
##
## LogLikelihood : -1581.975
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.6287
## Bayes 4.6815
## Shibata 4.6285
## Hannan-Quinn 4.6492
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 2.222 0.1361
## Lag[2*(p+q)+(p+q)-1][5] 2.746 0.6340
## Lag[4*(p+q)+(p+q)-1][9] 5.483 0.3504
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.9416 0.3319
## Lag[2*(p+q)+(p+q)-1][5] 2.8020 0.4444
## Lag[4*(p+q)+(p+q)-1][9] 4.8493 0.4519
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.8463 0.500 2.000 0.3576
## ARCH Lag[5] 1.3860 1.440 1.667 0.6228
## ARCH Lag[7] 1.8913 2.315 1.543 0.7403
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.5737
## Individual Statistics:
## mu 0.09826
## ma1 0.11526
## ma2 0.15323
## omega 0.13355
## alpha1 0.12522
## beta1 0.15516
## gamma1 0.34336
## shape 0.20223
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.89 2.11 2.59
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 1.3596 0.17439
## Negative Sign Bias 0.5869 0.55746
## Positive Sign Bias 2.0342 0.04232 **
## Joint Effect 5.4781 0.13995
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 45.37 0.0006077
## 2 30 48.50 0.0130290
## 3 40 65.66 0.0047850
## 4 50 66.20 0.0511737
##
##
## Elapsed time : 4.2954
OCB.a.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : apARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.077983 0.002197 -35.49424 0.000000
## ar1 0.128643 0.001395 92.22800 0.000000
## ar2 -0.982294 0.003826 -256.73113 0.000000
## ma1 -0.133474 0.001171 -113.98013 0.000000
## ma2 0.995942 0.000384 2596.00141 0.000000
## omega 0.157327 0.239951 0.65566 0.512041
## alpha1 0.139518 0.117908 1.18328 0.236697
## beta1 0.785605 0.176477 4.45160 0.000009
## gamma1 0.360098 0.346874 1.03812 0.299214
## delta 0.360201 0.456720 0.78867 0.430306
## shape 2.759141 0.524972 5.25579 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.077983 0.007694 -10.13576 0.000000
## ar1 0.128643 0.006564 19.59860 0.000000
## ar2 -0.982294 0.008294 -118.43224 0.000000
## ma1 -0.133474 0.005532 -24.12803 0.000000
## ma2 0.995942 0.000476 2092.42109 0.000000
## omega 0.157327 0.834883 0.18844 0.850530
## alpha1 0.139518 0.405279 0.34425 0.730656
## beta1 0.785605 0.590201 1.33108 0.183163
## gamma1 0.360098 0.981008 0.36707 0.713568
## delta 0.360201 1.665961 0.21621 0.828823
## shape 2.759141 1.477248 1.86776 0.061796
##
## LogLikelihood : -1386.57
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.0686
## Bayes 4.1412
## Shibata 4.0681
## Hannan-Quinn 4.0967
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.3697 0.5432
## Lag[2*(p+q)+(p+q)-1][11] 1.9591 1.0000
## Lag[4*(p+q)+(p+q)-1][19] 3.6513 0.9998
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.01228 0.9118
## Lag[2*(p+q)+(p+q)-1][5] 0.03515 0.9998
## Lag[4*(p+q)+(p+q)-1][9] 0.06084 1.0000
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.01462 0.500 2.000 0.9038
## ARCH Lag[5] 0.02545 1.440 1.667 0.9981
## ARCH Lag[7] 0.04312 2.315 1.543 0.9999
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 5.7781
## Individual Statistics:
## mu 2.49519
## ar1 3.48571
## ar2 0.64870
## ma1 3.23166
## ma2 0.32576
## omega 0.18445
## alpha1 0.26413
## beta1 0.22696
## gamma1 0.05807
## delta 0.24938
## shape 0.30649
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.49 2.75 3.27
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.5706 0.5685
## Negative Sign Bias 0.2940 0.7689
## Positive Sign Bias 0.2232 0.8234
## Joint Effect 0.7174 0.8691
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 25.40 0.1477
## 2 30 33.57 0.2554
## 3 40 45.05 0.2335
## 4 50 54.12 0.2853
##
##
## Elapsed time : 7.584002
SHB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(1,0,0)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.028448 0.059219 -0.48038 0.630955
## ar1 -0.028096 0.036918 -0.76105 0.446627
## omega 0.020278 0.010094 2.00889 0.044549
## alpha1 -0.072422 0.024353 -2.97388 0.002941
## beta1 0.984144 0.008163 120.56163 0.000000
## gamma1 0.161863 0.145703 1.11091 0.266607
## shape 4.119798 0.672083 6.12989 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.028448 0.086800 -0.32774 0.74311
## ar1 -0.028096 0.043109 -0.65174 0.51457
## omega 0.020278 0.030144 0.67270 0.50114
## alpha1 -0.072422 0.046842 -1.54611 0.12208
## beta1 0.984144 0.031714 31.03185 0.00000
## gamma1 0.161863 0.584618 0.27687 0.78188
## shape 4.119798 0.739458 5.57137 0.00000
##
## LogLikelihood : -1430.629
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.1852
## Bayes 4.2314
## Shibata 4.1850
## Hannan-Quinn 4.2031
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 2.967 0.08497
## Lag[2*(p+q)+(p+q)-1][2] 3.309 0.01883
## Lag[4*(p+q)+(p+q)-1][5] 3.760 0.26626
## d.o.f=1
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.2228 0.6369
## Lag[2*(p+q)+(p+q)-1][5] 0.8099 0.9010
## Lag[4*(p+q)+(p+q)-1][9] 1.4375 0.9605
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.3781 0.500 2.000 0.5386
## ARCH Lag[5] 0.4472 1.440 1.667 0.8991
## ARCH Lag[7] 0.8431 2.315 1.543 0.9377
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 2.5266
## Individual Statistics:
## mu 0.14058
## ar1 0.08129
## omega 0.34031
## alpha1 0.95023
## beta1 0.18996
## gamma1 0.11715
## shape 0.51051
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.700 0.48415
## Negative Sign Bias 0.148 0.88237
## Positive Sign Bias 1.915 0.05591 *
## Joint Effect 3.865 0.27645
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 56.55 1.341e-05
## 2 30 101.17 6.361e-10
## 3 40 132.04 4.882e-12
## 4 50 141.02 7.641e-11
##
##
## Elapsed time : 0.792836
SSB.a.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : apARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.021124 0.001796 -11.75937 0.000000
## ar1 -0.043266 0.002758 -15.68741 0.000000
## ar2 0.846448 0.010660 79.40280 0.000000
## ma1 0.045716 0.002200 20.77899 0.000000
## ma2 -0.817437 0.016138 -50.65386 0.000000
## omega 0.287525 0.125399 2.29288 0.021855
## alpha1 0.470001 0.295029 1.59307 0.111145
## beta1 0.532242 0.069369 7.67267 0.000000
## gamma1 0.229984 0.130846 1.75767 0.078804
## delta 0.359475 0.377154 0.95312 0.340527
## shape 2.115204 0.046045 45.93785 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.021124 0.018127 -1.165342 0.243880
## ar1 -0.043266 0.029238 -1.479798 0.138927
## ar2 0.846448 0.112337 7.534908 0.000000
## ma1 0.045716 0.022693 2.014497 0.043957
## ma2 -0.817437 0.174339 -4.688792 0.000003
## omega 0.287525 1.138103 0.252635 0.800550
## alpha1 0.470001 2.929237 0.160452 0.872525
## beta1 0.532242 0.330216 1.611799 0.107006
## gamma1 0.229984 0.216561 1.061980 0.288245
## delta 0.359475 3.823822 0.094009 0.925102
## shape 2.115204 0.070490 30.007302 0.000000
##
## LogLikelihood : -969.6714
##
## Information Criteria
## ------------------------------------
##
## Akaike 2.8549
## Bayes 2.9275
## Shibata 2.8544
## Hannan-Quinn 2.8830
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.4252 0.5144
## Lag[2*(p+q)+(p+q)-1][11] 5.2928 0.8827
## Lag[4*(p+q)+(p+q)-1][19] 11.4722 0.2520
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.3053 0.5806
## Lag[2*(p+q)+(p+q)-1][5] 0.8954 0.8829
## Lag[4*(p+q)+(p+q)-1][9] 1.2573 0.9730
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.05193 0.500 2.000 0.8197
## ARCH Lag[5] 0.54200 1.440 1.667 0.8712
## ARCH Lag[7] 0.73099 2.315 1.543 0.9530
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 8.8386
## Individual Statistics:
## mu 0.9468
## ar1 1.4621
## ar2 1.8242
## ma1 1.5216
## ma2 1.7244
## omega 0.1443
## alpha1 0.1634
## beta1 0.1226
## gamma1 1.8978
## delta 0.1738
## shape 0.1781
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.49 2.75 3.27
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.5173 0.6051
## Negative Sign Bias 0.5409 0.5887
## Positive Sign Bias 0.2635 0.7922
## Joint Effect 1.2978 0.7296
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 70.93 6.431e-08
## 2 30 95.66 4.776e-09
## 3 40 104.32 7.214e-08
## 4 50 112.34 6.995e-07
##
##
## Elapsed time : 8.820632
STB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(0,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043666 0.064952 0.67228 0.501404
## ma1 -0.076229 0.037721 -2.02088 0.043292
## ma2 -0.014466 0.038125 -0.37943 0.704371
## omega 0.029325 0.007042 4.16395 0.000031
## alpha1 -0.086533 0.022629 -3.82394 0.000131
## beta1 0.979441 0.002519 388.89088 0.000000
## gamma1 0.157671 0.036875 4.27583 0.000019
## shape 7.336517 2.242355 3.27179 0.001069
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043666 0.069488 0.62840 0.529744
## ma1 -0.076229 0.034981 -2.17916 0.029320
## ma2 -0.014466 0.039602 -0.36528 0.714903
## omega 0.029325 0.006041 4.85435 0.000001
## alpha1 -0.086533 0.025497 -3.39384 0.000689
## beta1 0.979441 0.000723 1353.86100 0.000000
## gamma1 0.157671 0.034702 4.54352 0.000006
## shape 7.336517 2.053239 3.57314 0.000353
##
## LogLikelihood : -1502.39
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.3971
## Bayes 4.4498
## Shibata 4.3968
## Hannan-Quinn 4.4175
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.01089 0.9169
## Lag[2*(p+q)+(p+q)-1][5] 1.04510 1.0000
## Lag[4*(p+q)+(p+q)-1][9] 2.80538 0.9179
## d.o.f=2
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 1.091 0.2962
## Lag[2*(p+q)+(p+q)-1][5] 3.171 0.3767
## Lag[4*(p+q)+(p+q)-1][9] 4.205 0.5545
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 2.837 0.500 2.000 0.09212
## ARCH Lag[5] 3.269 1.440 1.667 0.25305
## ARCH Lag[7] 3.540 2.315 1.543 0.41782
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.6937
## Individual Statistics:
## mu 0.07407
## ma1 0.16066
## ma2 0.44775
## omega 0.15811
## alpha1 0.31035
## beta1 0.25034
## gamma1 0.11846
## shape 0.50068
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.89 2.11 2.59
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.3048 0.7606
## Negative Sign Bias 0.5929 0.5535
## Positive Sign Bias 0.4375 0.6619
## Joint Effect 0.6470 0.8856
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 31.57 0.034888
## 2 30 38.11 0.119928
## 3 40 69.39 0.001953
## 4 50 57.47 0.190212
##
##
## Elapsed time : 1.57287
TCB.a.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : apARCH(1,1)
## Mean Model : ARFIMA(1,0,0)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043949 0.003788 11.6025 0.000000
## ar1 0.018905 0.000722 26.1665 0.000000
## omega 0.043058 0.024535 1.7550 0.079265
## alpha1 0.078491 0.022539 3.4824 0.000497
## beta1 0.917419 0.026088 35.1662 0.000000
## gamma1 0.579018 0.254555 2.2746 0.022928
## delta 0.479047 0.134607 3.5589 0.000372
## shape 3.068539 0.385643 7.9569 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.043949 0.002150 20.4377 0.000000
## ar1 0.018905 0.001042 18.1408 0.000000
## omega 0.043058 0.026873 1.6023 0.109099
## alpha1 0.078491 0.021715 3.6146 0.000301
## beta1 0.917419 0.025441 36.0610 0.000000
## gamma1 0.579018 0.275605 2.1009 0.035650
## delta 0.479047 0.132984 3.6023 0.000315
## shape 3.068539 0.487102 6.2996 0.000000
##
## LogLikelihood : -1412.207
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.1345
## Bayes 4.1873
## Shibata 4.1343
## Hannan-Quinn 4.1549
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.4657 0.4950
## Lag[2*(p+q)+(p+q)-1][2] 0.4803 0.9688
## Lag[4*(p+q)+(p+q)-1][5] 0.7148 0.9802
## d.o.f=1
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.002862 0.9573
## Lag[2*(p+q)+(p+q)-1][5] 0.008747 1.0000
## Lag[4*(p+q)+(p+q)-1][9] 0.014462 1.0000
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.002783 0.500 2.000 0.9579
## ARCH Lag[5] 0.007064 1.440 1.667 0.9997
## ARCH Lag[7] 0.010455 2.315 1.543 1.0000
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 4.4523
## Individual Statistics:
## mu 0.27163
## ar1 0.26830
## omega 0.09391
## alpha1 0.21481
## beta1 0.13728
## gamma1 0.04615
## delta 0.05201
## shape 0.34254
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.89 2.11 2.59
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.767776 0.4429
## Negative Sign Bias 0.198653 0.8426
## Positive Sign Bias 0.002455 0.9980
## Joint Effect 0.906208 0.8239
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 30.23 0.0488906
## 2 30 60.38 0.0005538
## 3 40 70.44 0.0015051
## 4 50 72.02 0.0177605
##
##
## Elapsed time : 3.901846
TPB.e.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : eGARCH(1,1)
## Mean Model : ARFIMA(0,0,1)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.019673 0.051556 0.38159 0.702765
## ma1 -0.089711 0.030364 -2.95458 0.003131
## omega 0.022844 0.005326 4.28874 0.000018
## alpha1 -0.038535 0.020158 -1.91164 0.055923
## beta1 0.985164 0.000150 6587.00296 0.000000
## gamma1 0.068198 0.020811 3.27711 0.001049
## shape 2.837310 0.326767 8.68299 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.019673 0.046340 0.42455 0.671168
## ma1 -0.089711 0.030740 -2.91844 0.003518
## omega 0.022844 0.005318 4.29528 0.000017
## alpha1 -0.038535 0.025805 -1.49332 0.135354
## beta1 0.985164 0.000176 5588.06486 0.000000
## gamma1 0.068198 0.014784 4.61286 0.000004
## shape 2.837310 0.356708 7.95414 0.000000
##
## LogLikelihood : -1415.167
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.1402
## Bayes 4.1864
## Shibata 4.1400
## Hannan-Quinn 4.1581
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.05202 0.8196
## Lag[2*(p+q)+(p+q)-1][2] 0.14012 0.9997
## Lag[4*(p+q)+(p+q)-1][5] 0.66810 0.9840
## d.o.f=1
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.03726 0.8469
## Lag[2*(p+q)+(p+q)-1][5] 0.15946 0.9952
## Lag[4*(p+q)+(p+q)-1][9] 0.29141 0.9997
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.09979 0.500 2.000 0.7521
## ARCH Lag[5] 0.18047 1.440 1.667 0.9700
## ARCH Lag[7] 0.27327 2.315 1.543 0.9941
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 0.8937
## Individual Statistics:
## mu 0.12511
## ma1 0.03769
## omega 0.12819
## alpha1 0.08534
## beta1 0.15607
## gamma1 0.12844
## shape 0.28771
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.1388 0.8897
## Negative Sign Bias 0.3082 0.7580
## Positive Sign Bias 0.8077 0.4196
## Joint Effect 0.8242 0.8437
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 29.48 0.058837
## 2 30 50.42 0.008128
## 3 40 65.20 0.005334
## 4 50 76.83 0.006737
##
##
## Elapsed time : 1.896572
VCB.g.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : gjrGARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.002356 0.043742 -0.053863 0.957044
## ar1 0.145742 0.018330 7.951082 0.000000
## ar2 -0.988104 0.008672 -113.944676 0.000000
## ma1 -0.150200 0.028917 -5.194144 0.000000
## ma2 0.963808 0.010670 90.331293 0.000000
## omega 0.000000 0.000188 0.000000 1.000000
## alpha1 0.000000 0.002528 0.000004 0.999996
## beta1 0.992934 0.000571 1739.532395 0.000000
## gamma1 0.010042 0.005229 1.920344 0.054814
## shape 4.256771 0.669250 6.360506 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.002356 0.046464 -0.050708 0.959558
## ar1 0.145742 0.027088 5.380207 0.000000
## ar2 -0.988104 0.008128 -121.565352 0.000000
## ma1 -0.150200 0.043602 -3.444764 0.000572
## ma2 0.963808 0.011011 87.532575 0.000000
## omega 0.000000 0.000009 0.000000 1.000000
## alpha1 0.000000 0.002968 0.000004 0.999997
## beta1 0.992934 0.000281 3537.976376 0.000000
## gamma1 0.010042 0.006491 1.546973 0.121870
## shape 4.256771 0.879952 4.837503 0.000001
##
## LogLikelihood : -1197.26
##
## Information Criteria
## ------------------------------------
##
## Akaike 3.5146
## Bayes 3.5806
## Shibata 3.5142
## Hannan-Quinn 3.5401
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.3564 0.5505
## Lag[2*(p+q)+(p+q)-1][11] 6.4538 0.2214
## Lag[4*(p+q)+(p+q)-1][19] 9.6759 0.5230
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.03058 0.8612
## Lag[2*(p+q)+(p+q)-1][5] 0.15745 0.9954
## Lag[4*(p+q)+(p+q)-1][9] 0.37312 0.9994
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.1101 0.500 2.000 0.7400
## ARCH Lag[5] 0.2252 1.440 1.667 0.9594
## ARCH Lag[7] 0.3349 2.315 1.543 0.9908
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.4203
## Individual Statistics:
## mu 0.16200
## ar1 0.08143
## ar2 0.15521
## ma1 0.06343
## ma2 0.26595
## omega 0.03608
## alpha1 0.07173
## beta1 0.08154
## gamma1 0.08177
## shape 0.06601
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.29 2.54 3.05
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.75227 0.4522
## Negative Sign Bias 0.05542 0.9558
## Positive Sign Bias 0.70771 0.4794
## Joint Effect 0.79771 0.8500
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 24.12 0.1916
## 2 30 36.89 0.1492
## 3 40 46.80 0.1828
## 4 50 46.55 0.5729
##
##
## Elapsed time : 2.052199
VIB.a.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : apARCH(1,1)
## Mean Model : ARFIMA(0,0,0)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu -0.088235 0.053889 -1.6373 0.101558
## omega 0.052071 0.022591 2.3050 0.021168
## alpha1 0.109703 0.028876 3.7991 0.000145
## beta1 0.875481 0.030312 28.8820 0.000000
## gamma1 0.575378 0.185707 3.0983 0.001946
## delta 0.316012 0.188856 1.6733 0.094269
## shape 3.545440 0.528705 6.7059 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu -0.088235 0.065346 -1.3503 0.176927
## omega 0.052071 0.026240 1.9844 0.047209
## alpha1 0.109703 0.027605 3.9740 0.000071
## beta1 0.875481 0.033911 25.8172 0.000000
## gamma1 0.575378 0.175563 3.2773 0.001048
## delta 0.316012 0.184304 1.7146 0.086413
## shape 3.545440 0.591085 5.9982 0.000000
##
## LogLikelihood : -1373.389
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.0186
## Bayes 4.0648
## Shibata 4.0184
## Hannan-Quinn 4.0365
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.5756 0.4480
## Lag[2*(p+q)+(p+q)-1][2] 0.6368 0.6327
## Lag[4*(p+q)+(p+q)-1][5] 1.1998 0.8130
## d.o.f=0
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.1819 0.6697
## Lag[2*(p+q)+(p+q)-1][5] 0.3052 0.9832
## Lag[4*(p+q)+(p+q)-1][9] 1.0920 0.9821
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 0.02360 0.500 2.000 0.8779
## ARCH Lag[5] 0.06899 1.440 1.667 0.9922
## ARCH Lag[7] 0.17873 2.315 1.543 0.9977
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.8966
## Individual Statistics:
## mu 0.2280
## omega 0.3085
## alpha1 0.3718
## beta1 0.3568
## gamma1 0.1720
## delta 0.3142
## shape 0.6772
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 1.69 1.9 2.35
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 1.4253 0.1545
## Negative Sign Bias 1.2243 0.2212
## Positive Sign Bias 0.3229 0.7469
## Joint Effect 3.6044 0.3075
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 59.35 4.908e-06
## 2 30 57.32 1.314e-03
## 3 40 102.00 1.540e-07
## 4 50 150.92 2.587e-12
##
##
## Elapsed time : 2.658072
VPB.s.11
##
## *---------------------------------*
## * GARCH Model Fit *
## *---------------------------------*
##
## Conditional Variance Dynamics
## -----------------------------------
## GARCH Model : sGARCH(1,1)
## Mean Model : ARFIMA(2,0,2)
## Distribution : std
##
## Optimal Parameters
## ------------------------------------
## Estimate Std. Error t value Pr(>|t|)
## mu 0.030599 0.052111 0.58719 0.557077
## ar1 0.590880 0.294681 2.00515 0.044947
## ar2 -0.612592 0.186657 -3.28192 0.001031
## ma1 -0.620068 0.313663 -1.97686 0.048057
## ma2 0.546779 0.210145 2.60191 0.009271
## omega 0.163665 0.098523 1.66118 0.096677
## alpha1 0.089035 0.035129 2.53454 0.011259
## beta1 0.887056 0.041499 21.37548 0.000000
## shape 3.496435 0.571099 6.12229 0.000000
##
## Robust Standard Errors:
## Estimate Std. Error t value Pr(>|t|)
## mu 0.030599 0.053165 0.57554 0.564925
## ar1 0.590880 0.354519 1.66671 0.095572
## ar2 -0.612592 0.184364 -3.32273 0.000891
## ma1 -0.620068 0.373255 -1.66125 0.096664
## ma2 0.546779 0.212745 2.57011 0.010166
## omega 0.163665 0.104463 1.56673 0.117178
## alpha1 0.089035 0.034350 2.59200 0.009542
## beta1 0.887056 0.044804 19.79874 0.000000
## shape 3.496435 0.562196 6.21925 0.000000
##
## LogLikelihood : -1381.264
##
## Information Criteria
## ------------------------------------
##
## Akaike 4.0473
## Bayes 4.1067
## Shibata 4.0470
## Hannan-Quinn 4.0703
##
## Weighted Ljung-Box Test on Standardized Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.3155 0.5743
## Lag[2*(p+q)+(p+q)-1][11] 4.8275 0.9800
## Lag[4*(p+q)+(p+q)-1][19] 7.8411 0.8157
## d.o.f=4
## H0 : No serial correlation
##
## Weighted Ljung-Box Test on Standardized Squared Residuals
## ------------------------------------
## statistic p-value
## Lag[1] 0.03208 0.8579
## Lag[2*(p+q)+(p+q)-1][5] 0.78218 0.9067
## Lag[4*(p+q)+(p+q)-1][9] 1.31744 0.9691
## d.o.f=2
##
## Weighted ARCH LM Tests
## ------------------------------------
## Statistic Shape Scale P-Value
## ARCH Lag[3] 1.048 0.500 2.000 0.3060
## ARCH Lag[5] 1.364 1.440 1.667 0.6288
## ARCH Lag[7] 1.436 2.315 1.543 0.8340
##
## Nyblom stability test
## ------------------------------------
## Joint Statistic: 1.7213
## Individual Statistics:
## mu 0.13847
## ar1 0.05920
## ar2 0.37246
## ma1 0.07306
## ma2 0.37026
## omega 0.27770
## alpha1 0.61437
## beta1 0.53096
## shape 0.43411
##
## Asymptotic Critical Values (10% 5% 1%)
## Joint Statistic: 2.1 2.32 2.82
## Individual Statistic: 0.35 0.47 0.75
##
## Sign Bias Test
## ------------------------------------
## t-value prob sig
## Sign Bias 0.4761 0.6342
## Negative Sign Bias 0.3637 0.7162
## Positive Sign Bias 0.1464 0.8837
## Joint Effect 0.4481 0.9301
##
##
## Adjusted Pearson Goodness-of-Fit Test:
## ------------------------------------
## group statistic p-value(g-1)
## 1 20 41.82 0.001874
## 2 30 36.54 0.158438
## 3 40 58.21 0.024541
## 4 50 71.59 0.019316
##
##
## Elapsed time : 0.590162
# ĐƯỜNG CONG MÔ TẢ MỨC ĐỘ BẤT ĐỐI XỨNG THÔNG TIN (NEWS IMPACT CURVE) #
plot(ACB.a.11, which=12)
plot(BAB.s.11, which=12)
plot(BID.g.11, which=12)
plot(CTG.e.11, which=12)
plot(EIB.s.11, which=12)
plot(HDB.e.11, which=12)
plot(MBB.e.11, which=12)
plot(MSB.e.11, which=12)
plot(NVB.e.11, which=12)
plot(OCB.a.11, which=12)
plot(SHB.e.11, which=12)
plot(SSB.a.11, which=12)
plot(STB.e.11, which=12)
plot(TCB.a.11, which=12)
plot(TPB.e.11, which=12)
plot(VCB.g.11, which=12)
plot(VIB.a.11, which=12)
plot(VPB.s.11, which=12)
# XÁC ĐỊNH PHÂN VỊ 5% CỦA PHÂN PHỐI STD #
ACB.res <- residuals(ACB.a.11)/sigma(ACB.a.11)
BAB.res <- residuals(BAB.s.11)/sigma(BAB.s.11)
BID.res <- residuals(BID.g.11)/sigma(BID.g.11)
CTG.res <- residuals(CTG.e.11)/sigma(CTG.e.11)
EIB.res <- residuals(EIB.s.11)/sigma(EIB.s.11)
HDB.res <- residuals(HDB.e.11)/sigma(HDB.e.11)
MBB.res <- residuals(MBB.e.11)/sigma(MBB.e.11)
MSB.res <- residuals(MSB.e.11)/sigma(MSB.e.11)
NVB.res <- residuals(NVB.e.11)/sigma(NVB.e.11)
OCB.res <- residuals(OCB.a.11)/sigma(OCB.a.11)
SHB.res <- residuals(SHB.e.11)/sigma(SHB.e.11)
SSB.res <- residuals(SSB.a.11)/sigma(SSB.a.11)
STB.res <- residuals(STB.e.11)/sigma(STB.e.11)
TCB.res <- residuals(TCB.a.11)/sigma(TCB.a.11)
TPB.res <- residuals(TPB.e.11)/sigma(TPB.e.11)
VCB.res <- residuals(VCB.g.11)/sigma(VCB.g.11)
VIB.res <- residuals(VIB.a.11)/sigma(VIB.a.11)
VPB.res <- residuals(VPB.s.11)/sigma(VPB.s.11)
ACB.df <- fitdist(distribution = "std", ACB.res, control = list())$pars
BAB.df <- fitdist(distribution = "std", BAB.res, control = list())$pars
BID.df <- fitdist(distribution = "std", BID.res, control = list())$pars
CTG.df <- fitdist(distribution = "std", CTG.res, control = list())$pars
EIB.df <- fitdist(distribution = "std", EIB.res, control = list())$pars
HDB.df <- fitdist(distribution = "std", HDB.res, control = list())$pars
MBB.df <- fitdist(distribution = "std", MBB.res, control = list())$pars
MSB.df <- fitdist(distribution = "std", MSB.res, control = list())$pars
NVB.df <- fitdist(distribution = "std", NVB.res, control = list())$pars
OCB.df <- fitdist(distribution = "std", OCB.res, control = list())$pars
SHB.df <- fitdist(distribution = "std", SHB.res, control = list())$pars
SSB.df <- fitdist(distribution = "std", SSB.res, control = list())$pars
STB.df <- fitdist(distribution = "std", STB.res, control = list())$pars
TCB.df <- fitdist(distribution = "std", TCB.res, control = list())$pars
TPB.df <- fitdist(distribution = "std", TPB.res, control = list())$pars
VCB.df <- fitdist(distribution = "std", VCB.res, control = list())$pars
VIB.df <- fitdist(distribution = "std", VIB.res, control = list())$pars
VPB.df <- fitdist(distribution = "std", VPB.res, control = list())$pars
ACB.std <- qstd(0.05, mean = ACB.df['mu'], sd = ACB.df['sigma'], nu = ACB.df['shape'])
BAB.std <- qstd(0.05, mean = BAB.df['mu'], sd = BAB.df['sigma'], nu = BAB.df['shape'])
BID.std <- qstd(0.05, mean = BID.df['mu'], sd = BID.df['sigma'], nu = BID.df['shape'])
CTG.std <- qstd(0.05, mean = CTG.df['mu'], sd = CTG.df['sigma'], nu = CTG.df['shape'])
EIB.std <- qstd(0.05, mean = EIB.df['mu'], sd = EIB.df['sigma'], nu = EIB.df['shape'])
HDB.std <- qstd(0.05, mean = HDB.df['mu'], sd = HDB.df['sigma'], nu = HDB.df['shape'])
MBB.std <- qstd(0.05, mean = MBB.df['mu'], sd = MBB.df['sigma'], nu = MBB.df['shape'])
MSB.std <- qstd(0.05, mean = MSB.df['mu'], sd = MSB.df['sigma'], nu = MSB.df['shape'])
NVB.std <- qstd(0.05, mean = NVB.df['mu'], sd = NVB.df['sigma'], nu = NVB.df['shape'])
OCB.std <- qstd(0.05, mean = OCB.df['mu'], sd = OCB.df['sigma'], nu = OCB.df['shape'])
SHB.std <- qstd(0.05, mean = SHB.df['mu'], sd = SHB.df['sigma'], nu = SHB.df['shape'])
SSB.std <- qstd(0.05, mean = SSB.df['mu'], sd = SSB.df['sigma'], nu = SSB.df['shape'])
STB.std <- qstd(0.05, mean = STB.df['mu'], sd = STB.df['sigma'], nu = STB.df['shape'])
TCB.std <- qstd(0.05, mean = TCB.df['mu'], sd = TCB.df['sigma'], nu = TCB.df['shape'])
TPB.std <- qstd(0.05, mean = TPB.df['mu'], sd = TPB.df['sigma'], nu = TPB.df['shape'])
VCB.std <- qstd(0.05, mean = VCB.df['mu'], sd = VCB.df['sigma'], nu = VCB.df['shape'])
VIB.std <- qstd(0.05, mean = VIB.df['mu'], sd = VIB.df['sigma'], nu = VIB.df['shape'])
VPB.std <- qstd(0.05, mean = VPB.df['mu'], sd = VPB.df['sigma'], nu = VPB.df['shape'])
# TÍNH GIÁ TRỊ VALUE AT RISK #
VaR.ACB<-ACB.std*ACB.a.11 @fit$sigma
VaR.BAB<-BAB.std*BAB.s.11 @fit$sigma
VaR.BID<-BID.std*BID.g.11 @fit$sigma
VaR.CTG<-CTG.std*CTG.e.11 @fit$sigma
VaR.EIB<-EIB.std*EIB.s.11 @fit$sigma
VaR.HDB<-HDB.std*HDB.e.11 @fit$sigma
VaR.MBB<-MBB.std*MBB.e.11 @fit$sigma
VaR.MSB<-MSB.std*MSB.e.11 @fit$sigma
VaR.NVB<-NVB.std*NVB.e.11 @fit$sigma
VaR.OCB<-OCB.std*OCB.a.11 @fit$sigma
VaR.SHB<-SHB.std*SHB.e.11 @fit$sigma
VaR.SSB<-SSB.std*SSB.a.11 @fit$sigma
VaR.STB<-STB.std*STB.e.11 @fit$sigma
VaR.TCB<-TCB.std*TCB.a.11 @fit$sigma
VaR.TPB<-TPB.std*TPB.e.11 @fit$sigma
VaR.VCB<-VCB.std*VCB.e.11 @fit$sigma
VaR.VIB<-VIB.std*VIB.a.11 @fit$sigma
VaR.VPB<-VPB.std*VPB.s.11 @fit$sigma
VaR <- data.frame(Date,VaR.ACB,VaR.BAB,VaR.BID,VaR.CTG,VaR.EIB,VaR.HDB,VaR.MBB,VaR.MSB,VaR.NVB,VaR.OCB,VaR.SHB,VaR.SSB,VaR.STB,VaR.TCB,VaR.TPB,VaR.VCB,VaR.VIB,VaR.VPB)
obj_list <- c("ACB", "BAB", "BID", "CTG", "EIB", "HDB", "MBB", "MSB", "NVB",
"OCB", "SHB", "SSB", "STB", "TCB", "TPB", "VCB", "VIB", "VPB")
for (obj_name in obj_list) {
var_column <- paste0("VaR.", obj_name)
p <- ggplot(VaR, aes(x = Date, y = get(var_column))) +
geom_line() +
labs(x = "", y = obj_name) +
ylim(min(VaR[[var_column]]), max(VaR[[var_column]]))
print(p)
}
