Indice de actividad económica El Salvador
library(tidyverse)## Warning: package 'tibble' was built under R version 4.0.5## Warning: package 'tidyr' was built under R version 4.0.5## Warning: package 'purrr' was built under R version 4.0.5## Warning: package 'dplyr' was built under R version 4.0.5## -- Attaching core tidyverse packages ------------------------ tidyverse 2.0.0 --
## v dplyr 1.0.8 v readr 2.1.4
## v forcats 1.0.0 v stringr 1.5.1
## v ggplot2 3.4.2 v tibble 3.1.6
## v lubridate 1.9.2 v tidyr 1.2.0
## v purrr 0.3.4
## -- Conflicts ------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
## i Use the [conflicted package](http://conflicted.r-lib.org/) to force all conflicts to become errorslibrary(forecast)## Warning: package 'forecast' was built under R version 4.0.5## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zooCARGA DE DATOS
df=readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/Indice de volumen de actividad economica/Índice_de_Volumen_de_la_Actividad_Económica_(IVAE)._Serie_desestacionalizada.xlsx")
desempleo_us= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/remesas/UNRATE.xls")
ipc_us= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/remesas/CPIAUCNS.xls")
remesas= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/pronosticos comercio exterior/Ingresos_mensuales_de_remesas_familiares - forecasted.xlsx")
impo_expo= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/pronosticos comercio exterior/Importaciones y exportaciones BCR.xlsx")
PIB_prod= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/PIB/PIB.xlsx")
IPC= readxl::read_excel("C:/Users/ronald.hernandez/Documents/pronosticos/IPC/Índice_de_Precios_al_Consumidor_(IPC).xlsx")
IPC_filtrado= IPC %>% filter(IPC$fecha>="2004-12-01")
impo_expo_filtrado=impo_expo %>% filter(fecha>="2004-12-01")
desempleo_us_filtrado=desempleo_us %>% filter(observation_date>="2004-12-01")
ipc_us_filtrado=ipc_us %>% filter(observation_date>="2004-12-01")
remesas_filtrado=remesas %>% filter(fecha>="2004-12-01")
PIB_filtrado=PIB_prod %>% filter(fecha >="2004-12-01")
serie_ivae <- ts(df$IVAE,
frequency = 12, start = c(2005,1))
autoplot(serie_ivae)
ANÁLISIS DE ESTACIONALIDAD
df_serie=data.frame(Y=as.matrix(serie_ivae), fecha=zoo::as.Date(time(serie_ivae)))
# Crear el resumen mensual y ordenar los meses
resumen_mensual <- df_serie %>%
mutate(mes = factor(month(fecha), labels = c("Ene", "Feb", "Mar", "Abr", "May", "Jun", "Jul", "Ago", "Sep", "Oct", "Nov", "Dic"), ordered = TRUE))
# Trazar el boxplot
ggplot(resumen_mensual, aes(x = mes, y = Y)) +
geom_boxplot() +
labs(x = "Mes", y = "IVAE", title = "Boxplot de IVAE El Salvador") 
autoplot(decompose(serie_ivae))
# DIVISIÓN DE DATOS
training <- window(serie_ivae, end =c(2022,5))
test <- window(serie_ivae, start =c(2022,6))MODELO MEDIA
modelo_media=Arima(training,order=c(0,1,0), include.mean=T, include.drift =T)
pronosticos=forecast(modelo_media)
autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_media)
##
## Ljung-Box test
##
## data: Residuals from ARIMA(0,1,0) with drift
## Q* = 33.733, df = 23, p-value = 0.06911
##
## Model df: 1. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set 0.0003820006 1.939076 1.229159 -0.01980926 1.288017 0.3412887
## Test set 1.1592467949 1.912226 1.458157 0.93768408 1.188892 0.4048724
## ACF1 Theil's U
## Training set 0.1126920 NA
## Test set 0.4552457 1.230682MODELO ARIMA
modelo_arima1=auto.arima(training, stepwise = F, d=1, D=1)
pronosticos=forecast(modelo_arima1)
autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_arima1)
##
## Ljung-Box test
##
## data: Residuals from ARIMA(0,1,1)(2,1,2)[12]
## Q* = 25.977, df = 19, p-value = 0.1308
##
## Model df: 5. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set -0.01485517 1.929518 1.260524 -0.04202314 1.307761 0.3499976
## Test set -1.87899957 3.019338 2.677522 -1.57280893 2.221545 0.7434417
## ACF1 Theil's U
## Training set 0.0183799 NA
## Test set 0.6405982 1.946319MODELO ARIMAX
reg=cbind("desempleo"=head(desempleo_us_filtrado$UNRATE, length(training)),
"importaciones_totales"=head(impo_expo_filtrado$`importaciones totales`, length(training)),
"exportaciones_totales"=head(impo_expo_filtrado$`exportaciones totales`, length(training)),
"PIB_producción"=head(PIB_filtrado$`Producto Interno Bruto trimestral por el enfoque de la producción`, length(training)))
new_xreg= cbind(head(desempleo_us_filtrado %>% filter(observation_date>"2022-05-01"),24) %>% pull(UNRATE),
head(impo_expo_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`importaciones totales`),
head(impo_expo_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`exportaciones totales`),
head(PIB_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`Producto Interno Bruto trimestral por el enfoque de la producción`))
modelo_arima2=auto.arima(training, xreg=reg, stepwise = F, d=1, D=1)
pronosticos=forecast(modelo_arima2, xreg=new_xreg)## Warning in forecast.forecast_ARIMA(modelo_arima2, xreg = new_xreg): xreg
## contains different column names from the xreg used in training. Please check
## that the regressors are in the same order.autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_arima2)
##
## Ljung-Box test
##
## data: Residuals from Regression with ARIMA(0,1,4)(0,1,1)[12] errors
## Q* = 18.912, df = 15, p-value = 0.2177
##
## Model df: 9. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set -0.04402747 1.244011 0.9134803 -0.05981333 0.9381635 0.2536372
## Test set 1.17760726 1.808699 1.5255622 0.95905522 1.2500179 0.4235881
## ACF1 Theil's U
## Training set 0.002435346 NA
## Test set 0.056695351 1.170672lmtest::coeftest(modelo_arima2)##
## z test of coefficients:
##
## Estimate Std. Error z value Pr(>|z|)
## ma1 -1.0615498 0.0899501 -11.8015 < 2.2e-16 ***
## ma2 0.0566141 0.1052900 0.5377 0.5907864
## ma3 -0.0855418 0.1186179 -0.7212 0.4708145
## ma4 0.1654842 0.0968217 1.7092 0.0874206 .
## sma1 -0.8782694 0.0635840 -13.8127 < 2.2e-16 ***
## desempleo -0.3280335 0.0978128 -3.3537 0.0007974 ***
## importaciones_totales -0.0025753 0.0011948 -2.1554 0.0311286 *
## exportaciones_totales 0.0114538 0.0037989 3.0150 0.0025696 **
## PIB_producción 0.7680350 0.0719547 10.6739 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1HOLT WINTERS
modelo_hw=forecast::hw(training)
pronosticos=forecast(modelo_hw)
autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_hw)
##
## Ljung-Box test
##
## data: Residuals from Holt-Winters' additive method
## Q* = 34.493, df = 8, p-value = 3.306e-05
##
## Model df: 16. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set 0.01743881 1.987038 1.354125 -0.002081692 1.419391 0.3759869
## Test set -2.53567772 3.141696 2.610154 -2.096924462 2.157298 0.7247363
## ACF1 Theil's U
## Training set 0.08763389 NA
## Test set 0.61518874 2.030137SUAVISADO EXPONENCIAL
modelo_ets=forecast::ets(training)
pronosticos=forecast(modelo_ets)
autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_ets)
##
## Ljung-Box test
##
## data: Residuals from ETS(A,N,N)
## Q* = 32.55, df = 22, p-value = 0.06849
##
## Model df: 2. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE ACF1
## Training set 0.1684727 1.955922 1.266075 0.1484995 1.325906 0.3515389 0.1013979
## Test set 3.4321803 4.065123 3.514672 2.7987893 2.869386 0.9758850 0.6455158
## Theil's U
## Training set NA
## Test set 2.63509TBATS
modelo_tbats=forecast::tbats(training)
pronosticos=forecast(modelo_tbats)
autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_tbats)
##
## Ljung-Box test
##
## data: Residuals from TBATS
## Q* = 23.135, df = 13, p-value = 0.04009
##
## Model df: 11. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set 0.1640077 1.828028 1.225426 0.1456915 1.283480 0.3402521
## Test set 5.8402993 6.414965 5.889981 4.7788024 4.821319 1.6354139
## ACF1 Theil's U
## Training set -0.08474781 NA
## Test set 0.68861823 4.173559MODELO DE REGRESIÓN LINEAL
reg=cbind("desempleo"=head(desempleo_us_filtrado$UNRATE, length(training)),
"importaciones_totales"=head(impo_expo_filtrado$`importaciones totales`, length(training)),
"exportaciones_totales"=head(impo_expo_filtrado$`exportaciones totales`, length(training)),
"PIB_producción"=head(PIB_filtrado$`Producto Interno Bruto trimestral por el enfoque de la producción`, length(training)),
"remesas"=head(remesas_filtrado$remesas, length(training)),
"IPC"=head(ipc_us_filtrado$CPIAUCNS, length(training)))
new_xreg= cbind(head(desempleo_us_filtrado %>% filter(observation_date>"2022-05-01"),24) %>% pull(UNRATE),
head(impo_expo_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`importaciones totales`),
head(impo_expo_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`exportaciones totales`),
head(PIB_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(`Producto Interno Bruto trimestral por el enfoque de la producción`),
head(remesas_filtrado %>% filter(fecha>"2022-05-01"),24) %>% pull(remesas),
head(ipc_us_filtrado %>% filter(observation_date>"2022-05-01"),24) %>% pull(CPIAUCNS))
modelo_arima2=Arima(training, xreg=reg, order=c(0,0,0))
pronosticos=forecast(modelo_arima2, xreg=new_xreg)## Warning in forecast.forecast_ARIMA(modelo_arima2, xreg = new_xreg): xreg
## contains different column names from the xreg used in training. Please check
## that the regressors are in the same order.autoplot(pronosticos)+ autolayer(test, series ="Test data")
checkresiduals(modelo_arima2)
##
## Ljung-Box test
##
## data: Residuals from Regression with ARIMA(0,0,0) errors
## Q* = 21.482, df = 17, p-value = 0.2055
##
## Model df: 7. Total lags used: 24accuracy(pronosticos, test)## ME RMSE MAE MPE MAPE MASE
## Training set 7.989357e-14 1.265689 0.9086107 -0.0182219 0.9424790 0.2522852
## Test set -3.059151e-01 1.385951 1.1091700 -0.2596493 0.9135611 0.3079725
## ACF1 Theil's U
## Training set -0.08746208 NA
## Test set 0.37004458 0.8907436lmtest::coeftest(modelo_arima2)##
## z test of coefficients:
##
## Estimate Std. Error z value Pr(>|z|)
## intercept 9.57047674 1.57916167 6.0605 1.357e-09 ***
## desempleo -0.28472443 0.07127590 -3.9947 6.478e-05 ***
## importaciones_totales -0.00293208 0.00120291 -2.4375 0.014789 *
## exportaciones_totales 0.00795400 0.00256523 3.1007 0.001931 **
## PIB_producción 0.80432653 0.04557942 17.6467 < 2.2e-16 ***
## remesas -0.00072222 0.00190420 -0.3793 0.704483
## IPC 0.04731461 0.02086051 2.2681 0.023321 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1