Importacion de datos
library(ggplot2)
library(forecast)
library(readxl)
ivae_ts <- read_excel("C:/Users/ariel/OneDrive/Documentos/IVAE Comercio, Transporte y Almacenamiento, Actividades de Alojamiento y de Servicio de Comidas.xlsx",
col_types = c("skip", "skip", "numeric"))
data=ivae_ts$IVAE %>% ts(start = c(2005,1),frequency = 12)->ivae
print(ivae)
## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2005 78.12 76.11 80.00 83.65 85.35 85.09 80.60 88.49 86.72 84.71
## 2006 84.88 79.37 81.41 77.13 90.74 92.21 83.44 87.59 92.79 94.95
## 2007 87.05 73.57 82.60 73.33 97.54 95.97 89.69 95.15 82.56 92.24
## 2008 82.84 79.22 77.77 82.72 88.08 85.69 79.21 85.79 91.35 95.24
## 2009 91.03 74.77 77.75 79.99 89.55 90.61 81.93 82.33 80.13 80.12
## 2010 84.63 78.01 87.11 82.61 91.15 85.22 79.99 83.35 80.51 82.77
## 2011 89.69 79.01 89.09 86.39 92.20 89.81 86.74 89.07 84.16 83.78
## 2012 87.51 83.03 90.75 85.98 101.29 97.31 89.30 93.15 87.54 90.93
## 2013 95.40 83.93 91.63 93.39 105.11 103.57 99.37 102.25 95.89 94.67
## 2014 100.91 93.64 103.55 98.38 108.55 105.44 98.82 99.84 93.74 87.65
## 2015 96.94 91.82 105.35 97.78 110.99 109.93 106.73 103.26 99.01 96.81
## 2016 100.21 96.88 103.62 104.05 115.49 116.91 107.19 108.80 103.93 101.97
## 2017 104.44 97.53 109.01 107.82 119.95 120.15 110.19 112.70 108.48 103.62
## 2018 109.97 99.26 112.96 111.43 123.45 123.92 111.82 115.17 111.43 108.02
## 2019 113.60 101.37 113.86 111.69 127.65 127.99 116.80 123.02
## Nov Dec
## 2005 97.57 101.84
## 2006 104.04 98.47
## 2007 101.61 102.62
## 2008 99.01 85.47
## 2009 88.77 99.54
## 2010 92.75 103.56
## 2011 92.23 102.59
## 2012 99.88 112.65
## 2013 102.26 118.51
## 2014 97.19 112.29
## 2015 105.85 119.40
## 2016 112.90 124.44
## 2017 118.82 125.87
## 2018 121.34 128.45
## 2019
autoplot(ivae,xlab="AƱos",ylab = "Indice", main="IVAE comercio, periodo 2005-2019 (agosto)"+theme_bw() )
2. proyeccion a seis meses
library(forecast)
modelo<-auto.arima(y=ivae)
summary(modelo)
## Series: ivae
## ARIMA(0,1,4)(2,1,1)[12]
##
## Coefficients:
## ma1 ma2 ma3 ma4 sar1 sar2 sma1
## -0.6118 -0.0413 -0.1670 0.0053 0.1274 0.1502 -0.7607
## s.e. 0.0807 0.0910 0.0967 0.0793 0.3208 0.2094 0.2988
##
## sigma^2 estimated as 17.13: log likelihood=-463
## AIC=942 AICc=942.94 BIC=966.75
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set 0.1039237 3.897014 2.77749 -0.06083988 2.984996 0.5851779
## ACF1
## Training set -0.001329976
pronostico<-forecast(modelo, h=6)
autoplot(pronostico)+xlab("AƱos")+ylab("Indice")+theme_bw()
library(forecast)
autoplot(pronostico$x,series="IVAE")+autolayer(pronostico$fitted, series="Pronostico")+ggtitle("Ajuste SARIMA")
3. Serie ampliada
ivae_h<-ts(as.numeric(rbind(as.matrix(pronostico$x), as.matrix(pronostico$mean))), start=c(2005,1), frequency=12)
print(ivae_h)
## Jan Feb Mar Apr May Jun Jul
## 2005 78.1200 76.1100 80.0000 83.6500 85.3500 85.0900 80.6000
## 2006 84.8800 79.3700 81.4100 77.1300 90.7400 92.2100 83.4400
## 2007 87.0500 73.5700 82.6000 73.3300 97.5400 95.9700 89.6900
## 2008 82.8400 79.2200 77.7700 82.7200 88.0800 85.6900 79.2100
## 2009 91.0300 74.7700 77.7500 79.9900 89.5500 90.6100 81.9300
## 2010 84.6300 78.0100 87.1100 82.6100 91.1500 85.2200 79.9900
## 2011 89.6900 79.0100 89.0900 86.3900 92.2000 89.8100 86.7400
## 2012 87.5100 83.0300 90.7500 85.9800 101.2900 97.3100 89.3000
## 2013 95.4000 83.9300 91.6300 93.3900 105.1100 103.5700 99.3700
## 2014 100.9100 93.6400 103.5500 98.3800 108.5500 105.4400 98.8200
## 2015 96.9400 91.8200 105.3500 97.7800 110.9900 109.9300 106.7300
## 2016 100.2100 96.8800 103.6200 104.0500 115.4900 116.9100 107.1900
## 2017 104.4400 97.5300 109.0100 107.8200 119.9500 120.1500 110.1900
## 2018 109.9700 99.2600 112.9600 111.4300 123.4500 123.9200 111.8200
## 2019 113.6000 101.3700 113.8600 111.6900 127.6500 127.9900 116.8000
## 2020 117.3646 107.4515
## Aug Sep Oct Nov Dec
## 2005 88.4900 86.7200 84.7100 97.5700 101.8400
## 2006 87.5900 92.7900 94.9500 104.0400 98.4700
## 2007 95.1500 82.5600 92.2400 101.6100 102.6200
## 2008 85.7900 91.3500 95.2400 99.0100 85.4700
## 2009 82.3300 80.1300 80.1200 88.7700 99.5400
## 2010 83.3500 80.5100 82.7700 92.7500 103.5600
## 2011 89.0700 84.1600 83.7800 92.2300 102.5900
## 2012 93.1500 87.5400 90.9300 99.8800 112.6500
## 2013 102.2500 95.8900 94.6700 102.2600 118.5100
## 2014 99.8400 93.7400 87.6500 97.1900 112.2900
## 2015 103.2600 99.0100 96.8100 105.8500 119.4000
## 2016 108.8000 103.9300 101.9700 112.9000 124.4400
## 2017 112.7000 108.4800 103.6200 118.8200 125.8700
## 2018 115.1700 111.4300 108.0200 121.3400 128.4500
## 2019 123.0200 116.0525 112.8630 124.7412 133.2311
## 2020
4. Descomposicion de la serie temporal
library(stats)
fit<-stl(ivae_h, "periodic")
autoplot(fit)+theme_bw()
TC<-fit$time.series[,2]
print(TC)
## Jan Feb Mar Apr May Jun Jul
## 2005 82.33599 82.88690 83.43781 83.98329 84.52878 85.02871 85.52865
## 2006 87.23083 87.36237 87.49392 87.85222 88.21052 88.56910 88.92768
## 2007 89.42136 89.34389 89.26641 89.19678 89.12714 89.32426 89.52137
## 2008 88.08709 87.52200 86.95692 86.80906 86.66120 86.62575 86.59030
## 2009 86.58525 86.10167 85.61808 85.01954 84.42099 84.31613 84.21126
## 2010 85.40349 85.38199 85.36048 85.40789 85.45530 85.60799 85.76068
## 2011 88.06382 88.38697 88.71012 88.74918 88.78824 88.68173 88.57523
## 2012 90.40029 90.84926 91.29823 91.78712 92.27601 92.78950 93.30299
## 2013 96.03940 96.69956 97.35973 97.83251 98.30530 98.75845 99.21159
## 2014 101.97014 101.91497 101.85980 101.45035 101.04091 100.45169 99.86247
## 2015 100.34152 101.04191 101.74231 102.40408 103.06585 103.44419 103.82252
## 2016 105.50758 106.03280 106.55803 107.06645 107.57488 107.91511 108.25535
## 2017 109.83593 110.25701 110.67810 111.04031 111.40252 111.65957 111.91661
## 2018 113.29753 113.65433 114.01113 114.31684 114.62255 114.82970 115.03686
## 2019 116.09741 116.65251 117.20762 117.70024 118.19285 118.43983 118.68682
## 2020 119.80032 119.89797
## Aug Sep Oct Nov Dec
## 2005 85.94977 86.37090 86.66767 86.96445 87.09764
## 2006 89.06870 89.20971 89.25312 89.29653 89.35895
## 2007 89.72318 89.92498 89.71278 89.50058 88.79383
## 2008 86.61695 86.64361 86.69512 86.74663 86.66594
## 2009 84.45514 84.69903 84.99033 85.28163 85.34256
## 2010 86.03410 86.30752 86.71742 87.12733 87.59557
## 2011 88.63773 88.70024 89.05294 89.40565 89.90297
## 2012 93.68599 94.06899 94.46935 94.86971 95.45455
## 2013 99.83081 100.45003 100.97977 101.50952 101.73983
## 2014 99.52643 99.19040 99.27922 99.36805 99.85478
## 2015 103.99123 104.15994 104.40127 104.64259 105.07508
## 2016 108.46056 108.66577 108.89665 109.12753 109.48173
## 2017 112.10011 112.28361 112.48267 112.68172 112.98963
## 2018 115.11895 115.20104 115.29513 115.38923 115.74332
## 2019 118.91963 119.15244 119.34168 119.53091 119.66562
## 2020
5. Calculo de las tasas
library(dplyr)
library(zoo)
TC%>%as.numeric()%>%as.data.frame()->TC_df
names(TC_df)<-c("TC")
TC_df %>% mutate(T_1_1=(TC/dplyr::lag(TC,n=1)-1)*100,
T_1_12=(TC/dplyr::lag(TC,n=12)-1)*100,
T_12_12=(rollapply(TC,12,mean,align='right',fill=NA)
/rollapply(dplyr::lag(TC,n=12),12,mean,align='right',fill=NA)-1)*100) %>%
#Aqui se realiza el centrado
mutate(T_1_12C=dplyr::lead(T_1_12,n = 6),
T_12_12C=dplyr::lead(T_12_12,n = 12)) %>% ts(start = c(2005,1),frequency = 12)->tabla_coyuntura
print(tail(tabla_coyuntura,n=12))
## TC T_1_1 T_1_12 T_12_12 T_1_12C T_12_12C
## Mar 2019 117.2076 0.4758645 2.803666 2.666980 3.430003 NA
## Apr 2019 117.7002 0.4202918 2.959668 2.668421 3.509726 NA
## May 2019 118.1929 0.4185327 3.114836 2.687916 3.589319 NA
## Jun 2019 118.4398 0.2089661 3.143899 2.714018 3.388792 NA
## Jul 2019 118.6868 0.2085303 3.172857 2.746685 3.189488 NA
## Aug 2019 118.9196 0.1961571 3.301522 2.797876 2.782157 NA
## Sep 2019 119.1524 0.1957731 3.430003 2.867479 NA NA
## Oct 2019 119.3417 0.1588191 3.509726 2.951606 NA NA
## Nov 2019 119.5309 0.1585672 3.589319 3.050187 NA NA
## Dec 2019 119.6656 0.1126920 3.388792 3.128981 NA NA
## Jan 2020 119.8003 0.1125652 3.189488 3.188100 NA NA
## Feb 2020 119.8980 0.0815116 2.782157 3.199073 NA NA
6. grafico de tasas(centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura %>% as.data.frame() %>% select(T_1_12C,T_12_12C) %>% ts(start=c(2005,1), frequency = 12)->tabla_coyuntura_graficos
autoplot(tabla_coyuntura_graficos)+xlab("AƱos")+ylab("Tasas porcentuales")+theme_bw()
