DATOS USCHANGE
#CARGAR DATOS
series<-uschange
?uschange
#OBTENER LA SERIE
autoplot(uschange)
autoplot(uschange[,1:2])
ts.plot(series[,1:2], xlab="Tiempo",col=c(1,2))
a <- VARselect(uschange[,1:2], lag.max=12,type="const")
a
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 5 1 1 5
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) -1.4387301 -1.4266106 -1.4581505 -1.4688220 -1.484526 -1.4626972
## HQ(n) -1.3947166 -1.3532547 -1.3554523 -1.3367815 -1.323143 -1.2719720
## SC(n) -1.3302232 -1.2457657 -1.2049676 -1.1433012 -1.086667 -0.9925004
## FPE(n) 0.2372304 0.2401289 0.2326861 0.2302383 0.226685 0.2317378
## 7 8 9 10 11 12
## AIC(n) -1.4475670 -1.4291678 -1.3889936 -1.3510597 -1.3180043 -1.2978215
## HQ(n) -1.2274995 -1.1797580 -1.1102414 -1.0429651 -0.9805674 -0.9310422
## SC(n) -0.9050323 -0.8142951 -0.7017830 -0.5915111 -0.4861177 -0.3935969
## FPE(n) 0.2353401 0.2398027 0.2497549 0.2595683 0.2684889 0.2742032
a$selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 5 1 1 5
#CREACION DEL MODELO
modelo1<-VAR(uschange[,1:2],p=1,type=c("const"))
modelo2<-VAR(uschange[,1:2],p=5,type=c("const"))
modelo1
##
## VAR Estimation Results:
## =======================
##
## Estimated coefficients for equation Consumption:
## ================================================
## Call:
## Consumption = Consumption.l1 + Income.l1 + const
##
## Consumption.l1 Income.l1 const
## 0.29645664 0.09434292 0.45810920
##
##
## Estimated coefficients for equation Income:
## ===========================================
## Call:
## Income = Consumption.l1 + Income.l1 + const
##
## Consumption.l1 Income.l1 const
## 0.5150933 -0.2547414 0.5146181
modelo2
##
## VAR Estimation Results:
## =======================
##
## Estimated coefficients for equation Consumption:
## ================================================
## Call:
## Consumption = Consumption.l1 + Income.l1 + Consumption.l2 + Income.l2 + Consumption.l3 + Income.l3 + Consumption.l4 + Income.l4 + Consumption.l5 + Income.l5 + const
##
## Consumption.l1 Income.l1 Consumption.l2 Income.l2 Consumption.l3
## 0.21855796 0.09445354 0.17212699 -0.03278838 0.28849780
## Income.l3 Consumption.l4 Income.l4 Consumption.l5 Income.l5
## -0.03818433 -0.02828834 -0.07767663 -0.03860699 -0.04324357
## const
## 0.35791445
##
##
## Estimated coefficients for equation Income:
## ===========================================
## Call:
## Income = Consumption.l1 + Income.l1 + Consumption.l2 + Income.l2 + Consumption.l3 + Income.l3 + Consumption.l4 + Income.l4 + Consumption.l5 + Income.l5 + const
##
## Consumption.l1 Income.l1 Consumption.l2 Income.l2 Consumption.l3
## 0.43750033 -0.29981380 0.04889127 -0.10080974 0.45234276
## Income.l3 Consumption.l4 Income.l4 Consumption.l5 Income.l5
## -0.14603080 0.30345938 -0.24189049 -0.07572447 -0.19026612
## const
## 0.53252727
aic1<-summary(modelo1)$logLik
aic1
## [1] -394.9332
aic2<-summary(modelo2)$logLik
aic2
## [1] -362.4965
#VALIDACION DEL MODELO
serial.test(modelo1, lags.pt=12, type="PT.asymptotic")
##
## Portmanteau Test (asymptotic)
##
## data: Residuals of VAR object modelo1
## Chi-squared = 52.272, df = 44, p-value = 0.1835
roots(modelo1)
## [1] 0.3737744 0.3320591
normality.test(modelo1, multivariate.only=FALSE)
## $Consumption
##
## JB-Test (univariate)
##
## data: Residual of Consumption equation
## Chi-squared = 47.401, df = 2, p-value = 5.094e-11
##
##
## $Income
##
## JB-Test (univariate)
##
## data: Residual of Income equation
## Chi-squared = 141.19, df = 2, p-value < 2.2e-16
##
##
## $JB
##
## JB-Test (multivariate)
##
## data: Residuals of VAR object modelo1
## Chi-squared = 178.11, df = 4, p-value < 2.2e-16
##
##
## $Skewness
##
## Skewness only (multivariate)
##
## data: Residuals of VAR object modelo1
## Chi-squared = 11.44, df = 2, p-value = 0.00328
##
##
## $Kurtosis
##
## Kurtosis only (multivariate)
##
## data: Residuals of VAR object modelo1
## Chi-squared = 166.67, df = 2, p-value < 2.2e-16
plot(modelo1, names="Income")
dev.off()
## null device
## 1
par(mar=c(1,1,1,1))
acf(residuals(modelo1)[,1])
pacf(residuals(modelo1)[,1])
modelo1$varresult$Income$coefficients
## Consumption.l1 Income.l1 const
## 0.5150933 -0.2547414 0.5146181
modelo1$varresult$Consumption$coefficients
## Consumption.l1 Income.l1 const
## 0.29645664 0.09434292 0.45810920
autoplot(forecast(modelo1))
DATOS MELSYD
#CARGAR DATOS
series<-melsyd
#OBTENER LA SERIE
autoplot(melsyd)
autoplot(melsyd[,1:3])
ts.plot(series[,1:3], xlab="Tiempo",col=c(1,3))
#HAY MUCHOS DATOS VACIOS POR LO QUE NO SE PUEDE REALIZAR
DATOS INSURANCE
#CARGAR DATOS
series<-insurance
#OBTENER LA SERIEç
autoplot(insurance)
autoplot(insurance[,1:2])
ts.plot(series[,1:2], xlab="Tiempo",col=c(1,2))
a <- VARselect(insurance[,1:2], lag.max=12,type="const")
a
## $selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 12 12 2 2
##
## $criteria
## 1 2 3 4 5 6
## AIC(n) -0.9415300 -1.2875411 -1.1802611 -1.0288969 -0.7804458 -0.5805128
## HQ(n) -0.8542583 -1.1420882 -0.9766270 -0.7670816 -0.4604494 -0.2023352
## SC(n) -0.6560576 -0.8117538 -0.5141588 -0.1724797 0.2662863 0.6565343
## FPE(n) 0.3906752 0.2780929 0.3139223 0.3746911 0.5009857 0.6527168
## 7 8 9 10 11 12
## AIC(n) -0.34407854 -0.43848191 -0.9023228 -1.3854606 -2.5780266 -2.7452042
## HQ(n) 0.09228024 0.05605805 -0.3496017 -0.7745583 -1.9089431 -2.0179396
## SC(n) 1.08328343 1.17919500 0.9056690 0.6128462 -0.3894049 -0.3662676
## FPE(n) 0.90988763 0.95168717 0.7328879 0.6104003 0.2955358 0.5636772
a$selection
## AIC(n) HQ(n) SC(n) FPE(n)
## 12 12 2 2
#CREACION DEL MODELO
modelo1<-VAR(insurance[,1:2],p=12,type=c("const"))
modelo2<-VAR(insurance[,1:2],p=2,type=c("const"))
modelo1
##
## VAR Estimation Results:
## =======================
##
## Estimated coefficients for equation Quotes:
## ===========================================
## Call:
## Quotes = Quotes.l1 + TV.advert.l1 + Quotes.l2 + TV.advert.l2 + Quotes.l3 + TV.advert.l3 + Quotes.l4 + TV.advert.l4 + Quotes.l5 + TV.advert.l5 + Quotes.l6 + TV.advert.l6 + Quotes.l7 + TV.advert.l7 + Quotes.l8 + TV.advert.l8 + Quotes.l9 + TV.advert.l9 + Quotes.l10 + TV.advert.l10 + Quotes.l11 + TV.advert.l11 + Quotes.l12 + TV.advert.l12 + const
##
## Quotes.l1 TV.advert.l1 Quotes.l2 TV.advert.l2 Quotes.l3
## 1.2470658 -2.4530452 1.0667824 -2.6162468 -0.3688445
## TV.advert.l3 Quotes.l4 TV.advert.l4 Quotes.l5 TV.advert.l5
## -0.2609116 0.5154976 -1.8976661 0.3596082 -1.3184915
## Quotes.l6 TV.advert.l6 Quotes.l7 TV.advert.l7 Quotes.l8
## 1.5935403 -2.1877329 -0.6528141 0.5197491 0.6483490
## TV.advert.l8 Quotes.l9 TV.advert.l9 Quotes.l10 TV.advert.l10
## -0.8583710 -0.2206740 0.0362382 0.7593211 -0.7383296
## Quotes.l11 TV.advert.l11 Quotes.l12 TV.advert.l12 const
## -1.2756525 1.1328161 -0.7724121 0.9547982 54.2172211
##
##
## Estimated coefficients for equation TV.advert:
## ==============================================
## Call:
## TV.advert = Quotes.l1 + TV.advert.l1 + Quotes.l2 + TV.advert.l2 + Quotes.l3 + TV.advert.l3 + Quotes.l4 + TV.advert.l4 + Quotes.l5 + TV.advert.l5 + Quotes.l6 + TV.advert.l6 + Quotes.l7 + TV.advert.l7 + Quotes.l8 + TV.advert.l8 + Quotes.l9 + TV.advert.l9 + Quotes.l10 + TV.advert.l10 + Quotes.l11 + TV.advert.l11 + Quotes.l12 + TV.advert.l12 + const
##
## Quotes.l1 TV.advert.l1 Quotes.l2 TV.advert.l2 Quotes.l3
## 0.2057646 -0.6306077 0.9308638 -1.7131892 -0.5956195
## TV.advert.l3 Quotes.l4 TV.advert.l4 Quotes.l5 TV.advert.l5
## 0.2971025 0.4408534 -1.1668884 0.2126698 -0.6784697
## Quotes.l6 TV.advert.l6 Quotes.l7 TV.advert.l7 Quotes.l8
## 0.8856756 -1.1119076 -0.6958322 0.5984264 0.7444307
## TV.advert.l8 Quotes.l9 TV.advert.l9 Quotes.l10 TV.advert.l10
## -0.6506325 -0.5900783 0.5505313 0.6171403 -0.7599420
## Quotes.l11 TV.advert.l11 Quotes.l12 TV.advert.l12 const
## -0.6373814 0.4255538 -0.3550255 0.5535360 27.9166420
modelo2
##
## VAR Estimation Results:
## =======================
##
## Estimated coefficients for equation Quotes:
## ===========================================
## Call:
## Quotes = Quotes.l1 + TV.advert.l1 + Quotes.l2 + TV.advert.l2 + const
##
## Quotes.l1 TV.advert.l1 Quotes.l2 TV.advert.l2 const
## 2.5587745 -2.7341322 -1.0943511 0.9337613 8.3490139
##
##
## Estimated coefficients for equation TV.advert:
## ==============================================
## Call:
## TV.advert = Quotes.l1 + TV.advert.l1 + Quotes.l2 + TV.advert.l2 + const
##
## Quotes.l1 TV.advert.l1 Quotes.l2 TV.advert.l2 const
## 1.1032431 -1.0161438 -0.6547575 0.6458306 5.0776769
aic1<-summary(modelo1)$logLik
aic1
## [1] 8.972301
aic2<-summary(modelo2)$logLik
aic2
## [1] -74.4654
serial.test(modelo2, lags.pt=12, type="PT.asymptotic")
##
## Portmanteau Test (asymptotic)
##
## data: Residuals of VAR object modelo2
## Chi-squared = 38.822, df = 40, p-value = 0.5232
roots(modelo2)
## [1] 0.7400142 0.7400142 0.6987953 0.2492407
normality.test(modelo2, multivariate.only=FALSE)
## $Quotes
##
## JB-Test (univariate)
##
## data: Residual of Quotes equation
## Chi-squared = 0.040082, df = 2, p-value = 0.9802
##
##
## $TV.advert
##
## JB-Test (univariate)
##
## data: Residual of TV.advert equation
## Chi-squared = 0.24206, df = 2, p-value = 0.886
##
##
## $JB
##
## JB-Test (multivariate)
##
## data: Residuals of VAR object modelo2
## Chi-squared = 37.953, df = 4, p-value = 1.146e-07
##
##
## $Skewness
##
## Skewness only (multivariate)
##
## data: Residuals of VAR object modelo2
## Chi-squared = 14.539, df = 2, p-value = 0.0006963
##
##
## $Kurtosis
##
## Kurtosis only (multivariate)
##
## data: Residuals of VAR object modelo2
## Chi-squared = 23.413, df = 2, p-value = 8.238e-06
autoplot(forecast(modelo2))
