Importar datos
library(ggplot2)
library(forecast)
library(readxl)
library(tidyr)
library(dplyr)
library(kableExtra)
serie_IMAE <- read_excel("serie_IMAE.xlsx", col_types = c("text", "numeric", "numeric", "numeric", "numeric", "numeric", "numeric"))1. El Salvador
1.1. Crear data de El Salvador
data_sv <- pivot_longer(data = serie_IMAE[,3],
cols = `El Salvador`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_sv <- data_sv[1:164,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_sv)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 86.73 80.85 87.19 83.92 91.42 93.46 86.39 86.72 87.57 85.27
## 2010 85.56 84.69 90.90 85.94 94.33 92.23 87.18 90.25 89.00 88.74
## 2011 90.27 86.73 94.32 90.79 98.50 97.59 92.16 94.22 92.33 89.06
## 2012 92.65 91.20 98.46 91.23 102.83 102.84 93.61 98.21 93.94 93.49
## 2013 95.67 90.77 96.12 96.34 103.08 101.58 96.42 98.96 97.74 96.22
## 2014 98.70 94.70 101.30 97.12 103.86 104.73 98.48 98.60 98.25 96.43
## 2015 98.87 94.82 103.15 98.75 105.65 105.45 101.67 101.06 100.64 100.44
## 2016 99.25 97.76 102.58 103.43 107.76 110.71 104.01 106.24 104.83 102.04
## 2017 101.41 98.97 108.44 101.40 110.85 113.63 105.51 107.88 106.21 103.28
## 2018 105.17 102.53 108.39 107.93 112.46 113.55 108.80 111.94 107.54 105.81
## 2019 108.10 106.41 113.02 109.95 114.95 114.86 111.24 113.28 111.66 108.32
## 2020 109.49 109.27 104.04 87.36 89.33 96.05 96.95 103.34 106.72 106.12
## 2021 106.84 107.04 114.51 109.72 115.43 115.19 112.16 114.23 113.82 109.73
## 2022 109.25 110.28 118.85 111.15 120.33 118.27 113.36 116.30
## Nov Dec
## 2009 91.86 99.64
## 2010 93.13 100.74
## 2011 96.86 103.91
## 2012 99.61 105.05
## 2013 101.24 108.37
## 2014 100.64 107.19
## 2015 104.90 109.86
## 2016 106.50 114.98
## 2017 110.39 117.56
## 2018 112.16 120.03
## 2019 116.10 122.08
## 2020 110.70 119.86
## 2021 116.70 123.69
## 2022library(ggplot2)
autoplot(imae_sv,
xlab = "años",
ylab = "Indice",
main = "IMAE total de El Salvador, periodo 2009 - 2022 (agosto)") +
theme_bw()
1.2. Proyección a seis meses
library(forecast)
modelo_sv <- auto.arima(y = imae_sv)
summary(modelo_sv)## Series: imae_sv
## ARIMA(2,0,0)(0,1,1)[12] with drift
##
## Coefficients:
## ar1 ar2 sma1 drift
## 0.9235 -0.1436 -0.8361 0.1618
## s.e. 0.0802 0.0807 0.0825 0.0199
##
## sigma^2 = 5.79: log likelihood = -354.8
## AIC=719.59 AICc=720 BIC=734.71
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set 0.06329184 2.2858 1.545284 0.03154945 1.523289 0.4028524
## ACF1
## Training set -0.003845263Se tiene:
\(ARIMA(2, 0, 0)(0, 1, 1)[12]\)
pronosticos_sv <- forecast(modelo_sv, h = 6)
autoplot(pronosticos_sv) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_sv$x, series = "IMAE El Salvador") + autolayer(pronosticos_sv$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
1.3. Serie ampliada
ivae_h_sv <- ts(as.numeric(rbind(as.matrix(pronosticos_sv$x),
as.matrix(pronosticos_sv$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_sv)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 86.7300 80.8500 87.1900 83.9200 91.4200 93.4600 86.3900 86.7200
## 2010 85.5600 84.6900 90.9000 85.9400 94.3300 92.2300 87.1800 90.2500
## 2011 90.2700 86.7300 94.3200 90.7900 98.5000 97.5900 92.1600 94.2200
## 2012 92.6500 91.2000 98.4600 91.2300 102.8300 102.8400 93.6100 98.2100
## 2013 95.6700 90.7700 96.1200 96.3400 103.0800 101.5800 96.4200 98.9600
## 2014 98.7000 94.7000 101.3000 97.1200 103.8600 104.7300 98.4800 98.6000
## 2015 98.8700 94.8200 103.1500 98.7500 105.6500 105.4500 101.6700 101.0600
## 2016 99.2500 97.7600 102.5800 103.4300 107.7600 110.7100 104.0100 106.2400
## 2017 101.4100 98.9700 108.4400 101.4000 110.8500 113.6300 105.5100 107.8800
## 2018 105.1700 102.5300 108.3900 107.9300 112.4600 113.5500 108.8000 111.9400
## 2019 108.1000 106.4100 113.0200 109.9500 114.9500 114.8600 111.2400 113.2800
## 2020 109.4900 109.2700 104.0400 87.3600 89.3300 96.0500 96.9500 103.3400
## 2021 106.8400 107.0400 114.5100 109.7200 115.4300 115.1900 112.1600 114.2300
## 2022 109.2500 110.2800 118.8500 111.1500 120.3300 118.2700 113.3600 116.3000
## 2023 113.6007 112.1216
## Sep Oct Nov Dec
## 2009 87.5700 85.2700 91.8600 99.6400
## 2010 89.0000 88.7400 93.1300 100.7400
## 2011 92.3300 89.0600 96.8600 103.9100
## 2012 93.9400 93.4900 99.6100 105.0500
## 2013 97.7400 96.2200 101.2400 108.3700
## 2014 98.2500 96.4300 100.6400 107.1900
## 2015 100.6400 100.4400 104.9000 109.8600
## 2016 104.8300 102.0400 106.5000 114.9800
## 2017 106.2100 103.2800 110.3900 117.5600
## 2018 107.5400 105.8100 112.1600 120.0300
## 2019 111.6600 108.3200 116.1000 122.0800
## 2020 106.7200 106.1200 110.7000 119.8600
## 2021 113.8200 109.7300 116.7000 123.6900
## 2022 115.3201 112.9454 118.8018 125.9326
## 20231.4. Descomposición de la serie temporal
library(stats)
fit_sv <- stl(ivae_h_sv,"periodic")
autoplot(fit_sv) + theme_bw()
TC_sv <- fit_sv$time.series[,2]
print(TC_sv)## Jan Feb Mar Apr May Jun Jul
## 2009 87.44441 87.60681 87.76921 87.92794 88.08667 88.25008 88.41348
## 2010 89.40028 89.55340 89.70651 89.86944 90.03236 90.22289 90.41342
## 2011 92.44334 92.80161 93.15988 93.39982 93.63975 93.84970 94.05965
## 2012 95.62347 95.95655 96.28962 96.54669 96.80376 96.94365 97.08353
## 2013 97.37067 97.57465 97.77863 98.01141 98.24418 98.49901 98.75383
## 2014 99.97696 100.08692 100.19688 100.19388 100.19088 100.13145 100.07202
## 2015 100.61674 100.90043 101.18412 101.48393 101.78373 101.98809 102.19245
## 2016 103.31134 103.67683 104.04231 104.37187 104.70143 104.96083 105.22023
## 2017 105.98874 106.19442 106.40010 106.64005 106.87999 107.13555 107.39110
## 2018 108.53384 108.78914 109.04444 109.27664 109.50885 109.71924 109.92962
## 2019 111.20285 111.44804 111.69322 111.95644 112.21966 112.39504 112.57042
## 2020 106.56927 105.30931 104.04934 103.38688 102.72442 102.59403 102.46364
## 2021 110.00438 110.88175 111.75911 112.26149 112.76387 113.08083 113.39780
## 2022 114.86324 115.06235 115.26147 115.47510 115.68873 115.83923 115.98974
## 2023 116.75131 116.86242
## Aug Sep Oct Nov Dec
## 2009 88.59338 88.77329 88.95274 89.13219 89.26623
## 2010 90.66268 90.91194 91.25980 91.60766 92.02550
## 2011 94.26976 94.47988 94.72779 94.97570 95.29959
## 2012 97.10042 97.11730 97.14252 97.16774 97.26920
## 2013 99.00009 99.24636 99.44622 99.64608 99.81152
## 2014 100.05130 100.03057 100.12173 100.21288 100.41481
## 2015 102.28931 102.38617 102.54615 102.70613 103.00874
## 2016 105.37231 105.52438 105.61158 105.69878 105.84376
## 2017 107.59635 107.80160 107.96728 108.13297 108.33340
## 2018 110.15601 110.38239 110.58903 110.79568 110.99926
## 2019 112.33843 112.10644 110.95596 109.80548 108.18737
## 2020 103.18295 103.90226 105.45812 107.01397 108.50918
## 2021 113.63685 113.87590 114.12496 114.37401 114.61862
## 2022 116.12786 116.26598 116.39214 116.51830 116.63480
## 20231.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_sv %>% as.numeric() %>% as.data.frame() -> TC_df_sv
names(TC_df_sv) <- c("TC_sv")
TC_df_sv %>% mutate(T_1_1 = (TC_sv/dplyr::lag(TC_sv, n = 1) - 1) * 100,
T_1_12 = (TC_sv/dplyr::lag(TC_sv, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_sv, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_sv, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_sv
tail(tabla_coyuntura_sv, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_sv | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 115.2615 | 0.17304748 | 3.133843 | 7.503904 | 2.098839 | NA |
| 115.4751 | 0.18534593 | 2.862610 | 7.009903 | 1.986578 | NA |
| 115.6887 | 0.18500303 | 2.593794 | 6.404204 | 1.874805 | NA |
| 115.8392 | 0.13009289 | 2.439316 | 5.758686 | 1.759034 | NA |
| 115.9897 | 0.12992386 | 2.285702 | 5.074966 | 1.643756 | NA |
| 116.1279 | 0.11907927 | 2.192074 | 4.432823 | 1.564430 | NA |
| 116.2660 | 0.11893764 | 2.098839 | 3.830585 | NA | NA |
| 116.3921 | 0.10850962 | 1.986578 | 3.328714 | NA | NA |
| 116.5183 | 0.10839200 | 1.874805 | 2.923816 | NA | NA |
| 116.6348 | 0.09999144 | 1.759034 | 2.608495 | NA | NA |
| 116.7513 | 0.09989156 | 1.643756 | 2.380675 | NA | NA |
| 116.8624 | 0.09516974 | 1.564430 | 2.198828 | NA | NA |
1.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_sv %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_sv
autoplot(tabla_coyuntura_graficos_sv) + theme_bw()
tabla_coyuntura_sv %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
2. Costa Rica
2.1. Crear data de Costa Rica
data_cr <- pivot_longer(data = serie_IMAE[,2],
cols = `Costa Rica`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_cr <- data_cr[1:165,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_cr)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 73.05 70.50 75.68 70.13 72.30 73.43 72.93 72.00 73.64 76.77
## 2010 75.10 73.53 79.92 73.27 75.74 76.43 76.13 75.58 77.14 79.74
## 2011 78.27 76.77 82.00 76.03 79.23 79.63 77.99 77.89 80.05 83.57
## 2012 82.37 82.95 86.03 78.55 82.23 81.83 80.60 81.77 82.75 85.69
## 2013 83.10 82.79 85.62 81.13 84.12 83.77 83.88 83.97 86.04 88.53
## 2014 86.41 87.04 89.12 83.12 86.04 85.36 86.63 86.17 88.14 92.55
## 2015 88.30 90.04 92.86 88.50 92.09 92.53 93.84 92.75 93.78 96.67
## 2016 94.53 95.60 96.36 93.13 95.39 95.66 94.94 94.84 98.12 101.26
## 2017 96.71 96.96 100.85 94.84 99.06 99.90 96.26 96.64 98.99 103.96
## 2018 99.21 99.00 103.55 99.62 104.59 103.43 101.46 101.10 101.62 106.09
## 2019 101.48 101.93 105.94 99.98 103.78 103.63 102.45 101.43 103.57 109.05
## 2020 102.20 104.23 102.60 89.65 91.81 95.78 91.86 92.51 97.39 101.72
## 2021 96.63 100.29 108.09 101.66 104.50 104.73 107.77 105.71 108.62 111.23
## 2022 106.31 108.14 117.49 105.61 108.90 109.10 110.06 110.34 110.48
## Nov Dec
## 2009 78.18 78.35
## 2010 82.16 81.06
## 2011 85.93 84.67
## 2012 89.26 88.63
## 2013 90.77 90.80
## 2014 94.00 95.23
## 2015 98.43 97.87
## 2016 103.90 103.79
## 2017 107.71 108.11
## 2018 108.90 108.01
## 2019 111.47 111.09
## 2020 105.12 110.61
## 2021 116.91 119.84
## 2022library(ggplot2)
autoplot(imae_cr,
xlab = "años",
ylab = "Indice",
main = "IMAE total de Costa Rica, periodo 2009 - 2022 (septiembre)") +
theme_bw()
2.2. Proyección a seis meses
library(forecast)
modelo_cr <- auto.arima(y = imae_cr)
summary(modelo_cr)## Series: imae_cr
## ARIMA(1,0,0)(0,1,1)[12] with drift
##
## Coefficients:
## ar1 sma1 drift
## 0.8511 -0.5096 0.2326
## s.e. 0.0419 0.0813 0.0423
##
## sigma^2 = 3.381: log likelihood = -311.14
## AIC=630.28 AICc=630.55 BIC=642.4
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set 0.01327525 1.753237 1.164367 -0.006606559 1.216715 0.28915
## ACF1
## Training set -0.068264Se tiene:
\(ARIMA(1, 0, 0)(0, 1, 1)[12]\)
pronosticos_cr <- forecast(modelo_cr, h = 6)
autoplot(pronosticos_cr) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_cr$x, series = "IMAE Costa Rica") + autolayer(pronosticos_cr$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
2.3. Serie ampliada
ivae_h_cr <- ts(as.numeric(rbind(as.matrix(pronosticos_cr$x),
as.matrix(pronosticos_cr$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_cr)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 73.0500 70.5000 75.6800 70.1300 72.3000 73.4300 72.9300 72.0000
## 2010 75.1000 73.5300 79.9200 73.2700 75.7400 76.4300 76.1300 75.5800
## 2011 78.2700 76.7700 82.0000 76.0300 79.2300 79.6300 77.9900 77.8900
## 2012 82.3700 82.9500 86.0300 78.5500 82.2300 81.8300 80.6000 81.7700
## 2013 83.1000 82.7900 85.6200 81.1300 84.1200 83.7700 83.8800 83.9700
## 2014 86.4100 87.0400 89.1200 83.1200 86.0400 85.3600 86.6300 86.1700
## 2015 88.3000 90.0400 92.8600 88.5000 92.0900 92.5300 93.8400 92.7500
## 2016 94.5300 95.6000 96.3600 93.1300 95.3900 95.6600 94.9400 94.8400
## 2017 96.7100 96.9600 100.8500 94.8400 99.0600 99.9000 96.2600 96.6400
## 2018 99.2100 99.0000 103.5500 99.6200 104.5900 103.4300 101.4600 101.1000
## 2019 101.4800 101.9300 105.9400 99.9800 103.7800 103.6300 102.4500 101.4300
## 2020 102.2000 104.2300 102.6000 89.6500 91.8100 95.7800 91.8600 92.5100
## 2021 96.6300 100.2900 108.0900 101.6600 104.5000 104.7300 107.7700 105.7100
## 2022 106.3100 108.1400 117.4900 105.6100 108.9000 109.1000 110.0600 110.3400
## 2023 108.5389 110.5827 117.3577
## Sep Oct Nov Dec
## 2009 73.6400 76.7700 78.1800 78.3500
## 2010 77.1400 79.7400 82.1600 81.0600
## 2011 80.0500 83.5700 85.9300 84.6700
## 2012 82.7500 85.6900 89.2600 88.6300
## 2013 86.0400 88.5300 90.7700 90.8000
## 2014 88.1400 92.5500 94.0000 95.2300
## 2015 93.7800 96.6700 98.4300 97.8700
## 2016 98.1200 101.2600 103.9000 103.7900
## 2017 98.9900 103.9600 107.7100 108.1100
## 2018 101.6200 106.0900 108.9000 108.0100
## 2019 103.5700 109.0500 111.4700 111.0900
## 2020 97.3900 101.7200 105.1200 110.6100
## 2021 108.6200 111.2300 116.9100 119.8400
## 2022 110.4800 113.9698 118.1828 120.7984
## 20232.4. Descomposición de la serie temporal
library(stats)
fit_cr <- stl(ivae_h_cr,"periodic")
autoplot(fit_cr) + theme_bw()
TC_cr <- fit_cr$time.series[,2]
print(TC_cr)## Jan Feb Mar Apr May Jun Jul
## 2009 73.49623 73.61103 73.72584 73.84305 73.96026 74.08960 74.21893
## 2010 75.56195 75.89689 76.23184 76.55175 76.87166 77.11590 77.36013
## 2011 78.63026 78.90632 79.18239 79.48676 79.79114 80.10348 80.41582
## 2012 82.17225 82.44060 82.70896 82.94258 83.17620 83.33039 83.48458
## 2013 84.20802 84.44128 84.67455 84.90429 85.13404 85.36738 85.60071
## 2014 86.87680 87.08175 87.28670 87.54434 87.80199 88.07168 88.34138
## 2015 90.79669 91.34148 91.88626 92.34790 92.80954 93.19650 93.58347
## 2016 95.28402 95.56406 95.84410 96.22755 96.61099 96.97056 97.33012
## 2017 98.74610 98.91413 99.08216 99.28649 99.49082 99.76547 100.04012
## 2018 101.95599 102.29784 102.63969 102.84284 103.04600 103.18399 103.32198
## 2019 103.65880 103.77934 103.89987 104.09372 104.28756 104.48074 104.67393
## 2020 101.93479 101.06362 100.19245 99.50019 98.80793 98.45018 98.09244
## 2021 102.02720 103.05190 104.07659 105.00632 105.93605 106.76990 107.60375
## 2022 110.65032 110.86943 111.08854 111.25684 111.42513 111.57356 111.72198
## 2023 112.87662 113.08235 113.28809
## Aug Sep Oct Nov Dec
## 2009 74.37561 74.53230 74.75538 74.97846 75.27020
## 2010 77.54315 77.72618 77.93566 78.14514 78.38770
## 2011 80.72948 81.04313 81.34108 81.63902 81.90564
## 2012 83.55125 83.61792 83.73136 83.84479 84.02640
## 2013 85.85098 86.10126 86.31890 86.53655 86.70667
## 2014 88.63204 88.92271 89.33594 89.74918 90.27294
## 2015 93.92490 94.26634 94.55107 94.83581 95.05991
## 2016 97.58036 97.83059 98.07082 98.31104 98.52857
## 2017 100.29855 100.55698 100.87001 101.18305 101.56952
## 2018 103.42381 103.52565 103.55486 103.58407 103.62144
## 2019 104.64510 104.61627 104.10928 103.60229 102.76854
## 2020 98.26294 98.43344 99.18844 99.94343 100.98532
## 2021 108.31461 109.02546 109.50710 109.98874 110.31953
## 2022 111.90155 112.08111 112.27589 112.47066 112.67364
## 20232.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_cr %>% as.numeric() %>% as.data.frame() -> TC_df_cr
names(TC_df_cr) <- c("TC_cr")
TC_df_cr %>% mutate(T_1_1 = (TC_cr/dplyr::lag(TC_cr, n = 1) - 1) * 100,
T_1_12 = (TC_cr/dplyr::lag(TC_cr, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_cr, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_cr, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_cr
tail(tabla_coyuntura_cr, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_cr | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 111.2568 | 0.1514999 | 5.952514 | 8.704458 | 2.528406 | NA |
| 111.4251 | 0.1512707 | 5.181510 | 8.518254 | 2.256520 | NA |
| 111.5736 | 0.1332048 | 4.499075 | 8.172465 | 2.133900 | NA |
| 111.7220 | 0.1330276 | 3.827216 | 7.671330 | 2.012013 | NA |
| 111.9015 | 0.1607250 | 3.311593 | 7.088425 | 1.995973 | NA |
| 112.0811 | 0.1604671 | 2.802693 | 6.426472 | 1.979996 | NA |
| 112.2759 | 0.1737787 | 2.528406 | 5.775807 | NA | NA |
| 112.4707 | 0.1734772 | 2.256520 | 5.135735 | NA | NA |
| 112.6736 | 0.1804743 | 2.133900 | 4.553702 | NA | NA |
| 112.8766 | 0.1801492 | 2.012013 | 4.027487 | NA | NA |
| 113.0824 | 0.1822634 | 1.995973 | 3.571466 | NA | NA |
| 113.2881 | 0.1819318 | 1.979996 | 3.183534 | NA | NA |
2.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_cr %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_cr
autoplot(tabla_coyuntura_graficos_cr) + theme_bw()
tabla_coyuntura_cr %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
3. Guatemala
3.1. Crear data de Guatemala
data_gtm <- pivot_longer(data = serie_IMAE[,4],
cols = `Guatemala`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_gtm <- data_gtm[1:165,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_gtm)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 86.65 84.95 90.17 87.79 85.69 83.92 87.09 85.90 84.65 87.06
## 2010 88.43 87.09 94.14 89.68 88.28 87.49 88.03 87.35 86.92 88.69
## 2011 92.16 91.28 96.96 93.60 92.20 91.60 92.65 92.61 92.08 91.78
## 2012 95.05 94.95 101.10 95.13 95.58 94.13 94.97 95.31 94.02 96.32
## 2013 99.07 98.81 101.72 101.20 99.50 96.72 98.64 98.67 97.72 99.48
## 2014 102.75 102.57 106.76 104.80 104.40 101.05 103.78 102.20 101.78 103.90
## 2015 107.76 107.15 111.74 107.66 106.67 105.63 108.72 107.53 106.64 108.45
## 2016 109.74 109.44 112.96 112.29 111.12 108.40 109.35 110.41 109.80 110.43
## 2017 115.42 114.30 118.07 114.70 113.72 111.63 113.82 113.93 112.07 113.68
## 2018 117.75 117.77 121.77 119.59 118.71 116.35 118.22 118.04 115.42 117.98
## 2019 122.08 122.76 126.05 123.95 123.67 120.45 122.93 121.94 120.78 122.99
## 2020 127.01 125.51 121.38 112.73 111.49 111.55 118.50 120.60 121.73 125.20
## 2021 128.88 128.61 133.29 130.06 130.01 127.53 131.22 130.13 128.77 130.62
## 2022 134.95 134.14 139.23 135.83 135.53 132.00 135.09 136.00 133.97
## Nov Dec
## 2009 87.94 95.19
## 2010 91.35 98.92
## 2011 95.86 101.43
## 2012 98.92 104.11
## 2013 102.16 106.30
## 2014 107.09 112.27
## 2015 111.44 115.24
## 2016 114.99 120.63
## 2017 116.91 122.56
## 2018 121.04 125.20
## 2019 126.94 130.45
## 2020 128.05 135.04
## 2021 135.34 140.77
## 2022library(ggplot2)
autoplot(imae_gtm,
xlab = "años",
ylab = "Indice",
main = "IMAE total de Guatemala, periodo 2009 - 2022 (septiembre)") +
theme_bw()
3.2. Proyección a seis meses
library(forecast)
modelo_gtm <- auto.arima(y = imae_gtm)
summary(modelo_gtm)## Series: imae_gtm
## ARIMA(1,0,1)(0,1,1)[12] with drift
##
## Coefficients:
## ar1 ma1 sma1 drift
## 0.7685 0.2206 -0.7846 0.3047
## s.e. 0.0640 0.1040 0.0732 0.0147
##
## sigma^2 = 1.948: log likelihood = -272.4
## AIC=554.8 AICc=555.2 BIC=569.95
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set -0.02030934 1.326456 0.8825237 -0.03497516 0.8008833 0.2039824
## ACF1
## Training set 0.003115459Se tiene:
\(ARIMA(1, 0, 1)(0, 1, 1)[12]\)
pronosticos_gtm <- forecast(modelo_gtm, h = 6)
autoplot(pronosticos_gtm) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_gtm$x, series = "IMAE Guatemala") + autolayer(pronosticos_gtm$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
3.3. Serie ampliada
ivae_h_gtm <- ts(as.numeric(rbind(as.matrix(pronosticos_gtm$x),
as.matrix(pronosticos_gtm$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_gtm)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 86.6500 84.9500 90.1700 87.7900 85.6900 83.9200 87.0900 85.9000
## 2010 88.4300 87.0900 94.1400 89.6800 88.2800 87.4900 88.0300 87.3500
## 2011 92.1600 91.2800 96.9600 93.6000 92.2000 91.6000 92.6500 92.6100
## 2012 95.0500 94.9500 101.1000 95.1300 95.5800 94.1300 94.9700 95.3100
## 2013 99.0700 98.8100 101.7200 101.2000 99.5000 96.7200 98.6400 98.6700
## 2014 102.7500 102.5700 106.7600 104.8000 104.4000 101.0500 103.7800 102.2000
## 2015 107.7600 107.1500 111.7400 107.6600 106.6700 105.6300 108.7200 107.5300
## 2016 109.7400 109.4400 112.9600 112.2900 111.1200 108.4000 109.3500 110.4100
## 2017 115.4200 114.3000 118.0700 114.7000 113.7200 111.6300 113.8200 113.9300
## 2018 117.7500 117.7700 121.7700 119.5900 118.7100 116.3500 118.2200 118.0400
## 2019 122.0800 122.7600 126.0500 123.9500 123.6700 120.4500 122.9300 121.9400
## 2020 127.0100 125.5100 121.3800 112.7300 111.4900 111.5500 118.5000 120.6000
## 2021 128.8800 128.6100 133.2900 130.0600 130.0100 127.5300 131.2200 130.1300
## 2022 134.9500 134.1400 139.2300 135.8300 135.5300 132.0000 135.0900 136.0000
## 2023 137.9819 137.2163 140.3554
## Sep Oct Nov Dec
## 2009 84.6500 87.0600 87.9400 95.1900
## 2010 86.9200 88.6900 91.3500 98.9200
## 2011 92.0800 91.7800 95.8600 101.4300
## 2012 94.0200 96.3200 98.9200 104.1100
## 2013 97.7200 99.4800 102.1600 106.3000
## 2014 101.7800 103.9000 107.0900 112.2700
## 2015 106.6400 108.4500 111.4400 115.2400
## 2016 109.8000 110.4300 114.9900 120.6300
## 2017 112.0700 113.6800 116.9100 122.5600
## 2018 115.4200 117.9800 121.0400 125.2000
## 2019 120.7800 122.9900 126.9400 130.4500
## 2020 121.7300 125.2000 128.0500 135.0400
## 2021 128.7700 130.6200 135.3400 140.7700
## 2022 133.9700 135.3104 138.3894 143.2999
## 20233.4. Descomposición de la serie temporal
library(stats)
fit_gtm <- stl(ivae_h_gtm,"periodic")
autoplot(fit_gtm) + theme_bw()
TC_gtm <- fit_gtm$time.series[,2]
print(TC_gtm)## Jan Feb Mar Apr May Jun Jul
## 2009 86.06129 86.27892 86.49655 86.70443 86.91230 87.11913 87.32597
## 2010 88.53167 88.71821 88.90476 89.11966 89.33456 89.59635 89.85814
## 2011 91.78800 92.20524 92.62247 92.98003 93.33759 93.61300 93.88842
## 2012 95.24801 95.49523 95.74245 96.01742 96.29239 96.56241 96.83243
## 2013 98.52919 98.83917 99.14916 99.43305 99.71695 99.97579 100.23463
## 2014 102.24919 102.65525 103.06131 103.46417 103.86703 104.26639 104.66576
## 2015 106.68770 107.10443 107.52117 107.90686 108.29256 108.56870 108.84484
## 2016 110.11261 110.33241 110.55220 110.83620 111.12021 111.49918 111.87815
## 2017 113.84974 114.12818 114.40662 114.61404 114.82146 115.00388 115.18630
## 2018 117.20538 117.60032 117.99525 118.33719 118.67913 118.96821 119.25729
## 2019 121.38235 121.83153 122.28072 122.72953 123.17835 123.55066 123.92297
## 2020 121.49607 121.07629 120.65651 120.62134 120.58617 120.91099 121.23580
## 2021 127.78761 128.59039 129.39317 129.96080 130.52842 131.02954 131.53065
## 2022 134.24207 134.65968 135.07729 135.43940 135.80152 136.03393 136.26633
## 2023 137.15706 137.27954 137.40203
## Aug Sep Oct Nov Dec
## 2009 87.53758 87.74919 87.94115 88.13311 88.33239
## 2010 90.13509 90.41205 90.71710 91.02214 91.40507
## 2011 94.12665 94.36487 94.58231 94.79975 95.02388
## 2012 97.08568 97.33894 97.62283 97.90672 98.21795
## 2013 100.50652 100.77840 101.11729 101.45618 101.85268
## 2014 105.01319 105.36063 105.66678 105.97292 106.33031
## 2015 109.04041 109.23598 109.45280 109.66961 109.89111
## 2016 112.25857 112.63899 112.95127 113.26354 113.55664
## 2017 115.43389 115.68149 116.03494 116.38840 116.79689
## 2018 119.55196 119.84663 120.19984 120.55304 120.96770
## 2019 123.95542 123.98787 123.46096 122.93405 122.21506
## 2020 122.05656 122.87733 124.14347 125.40962 126.59862
## 2021 132.00831 132.48597 132.93681 133.38764 133.81486
## 2022 136.42481 136.58328 136.73331 136.88334 137.02020
## 20233.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_gtm %>% as.numeric() %>% as.data.frame() -> TC_df_gtm
names(TC_df_gtm) <- c("TC_gtm")
TC_df_gtm %>% mutate(T_1_1 = (TC_gtm/dplyr::lag(TC_gtm, n = 1) - 1) * 100,
T_1_12 = (TC_gtm/dplyr::lag(TC_gtm, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_gtm, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_gtm, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_gtm
tail(tabla_coyuntura_gtm, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_gtm | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 135.4394 | 0.26808085 | 4.215585 | 6.507957 | 2.855871 | NA |
| 135.8015 | 0.26736409 | 4.039809 | 6.155774 | 2.620706 | NA |
| 136.0339 | 0.17113588 | 3.819285 | 5.778241 | 2.395356 | NA |
| 136.2663 | 0.17084351 | 3.600441 | 5.375992 | 2.171441 | NA |
| 136.4248 | 0.11629904 | 3.345620 | 4.981783 | 1.945544 | NA |
| 136.5833 | 0.11616394 | 3.092637 | 4.595197 | 1.721043 | NA |
| 136.7333 | 0.10984309 | 2.855871 | 4.248599 | NA | NA |
| 136.8833 | 0.10972257 | 2.620706 | 3.940774 | NA | NA |
| 137.0202 | 0.09998366 | 2.395356 | 3.667790 | NA | NA |
| 137.1571 | 0.09988379 | 2.171441 | 3.428776 | NA | NA |
| 137.2795 | 0.08930141 | 1.945544 | 3.198159 | NA | NA |
| 137.4020 | 0.08922174 | 1.721043 | 2.975677 | NA | NA |
3.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_gtm %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_gtm
autoplot(tabla_coyuntura_graficos_gtm) + theme_bw()
tabla_coyuntura_gtm %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
4. Honduras
4.1. Crear data de Honduras
data_hnd <- pivot_longer(data = serie_IMAE[,5],
cols = `Honduras`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_hnd <- data_hnd[1:164,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_hnd)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 157.26 159.33 169.91 156.18 164.17 163.04 155.42 159.89 157.82 166.33
## 2010 165.28 166.91 179.91 165.46 173.89 171.00 162.53 166.65 175.18 172.00
## 2011 176.96 179.46 190.71 175.18 184.30 182.33 175.83 185.67 182.03 185.82
## 2012 181.51 189.25 202.52 183.81 193.45 192.28 185.89 193.61 188.79 199.97
## 2013 189.68 192.66 196.37 195.49 199.00 194.38 190.45 196.66 191.32 201.79
## 2014 194.20 197.58 205.41 197.36 207.03 198.09 194.18 199.21 197.73 205.50
## 2015 200.82 202.02 214.06 206.39 206.66 206.13 201.94 207.78 204.91 213.81
## 2016 207.87 210.56 220.51 211.07 214.45 216.00 205.61 215.98 212.31 220.76
## 2017 219.37 221.50 233.93 218.03 225.53 225.90 216.75 229.08 226.26 232.75
## 2018 228.97 228.12 237.11 227.12 234.88 234.03 225.04 238.66 232.55 244.93
## 2019 235.30 235.08 246.40 234.80 241.51 235.46 238.02 244.65 239.69 252.72
## 2020 242.49 241.65 218.27 186.88 189.07 208.71 209.30 225.80 230.24 249.34
## 2021 229.97 236.28 251.05 235.96 242.36 247.40 239.81 256.77 246.87 265.45
## 2022 247.27 246.62 263.70 248.77 254.73 256.23 246.20 272.14
## Nov Dec
## 2009 163.97 176.16
## 2010 175.48 186.89
## 2011 188.18 198.66
## 2012 199.48 203.10
## 2013 201.54 213.57
## 2014 203.26 221.72
## 2015 214.73 231.40
## 2016 227.59 245.58
## 2017 235.80 251.23
## 2018 245.16 262.48
## 2019 250.26 273.80
## 2020 218.89 258.08
## 2021 264.73 279.05
## 2022library(ggplot2)
autoplot(imae_hnd,
xlab = "años",
ylab = "Indice",
main = "IMAE total de Honduras, periodo 2009 - 2022 (agosto)") +
theme_bw()
4.2. Proyección a seis meses
library(forecast)
modelo_hnd <- auto.arima(y = imae_hnd)
summary(modelo_hnd)## Series: imae_hnd
## ARIMA(1,0,0)(0,1,1)[12] with drift
##
## Coefficients:
## ar1 sma1 drift
## 0.7873 -0.6826 0.5907
## s.e. 0.0505 0.0664 0.0756
##
## sigma^2 = 40.52: log likelihood = -499.71
## AIC=1007.42 AICc=1007.69 BIC=1019.51
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set 0.1146119 6.067638 3.641689 0.003617344 1.685864 0.3224686
## ACF1
## Training set -0.03097548Se tiene:
\(ARIMA(1, 0, 0)(0, 1, 1)[12]\)
pronosticos_hnd <- forecast(modelo_hnd, h = 6)
autoplot(pronosticos_hnd) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_hnd$x, series = "IMAE Honduras") + autolayer(pronosticos_hnd$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
4.3. Serie ampliada
ivae_h_hnd <- ts(as.numeric(rbind(as.matrix(pronosticos_hnd$x),
as.matrix(pronosticos_hnd$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_hnd)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 157.2600 159.3300 169.9100 156.1800 164.1700 163.0400 155.4200 159.8900
## 2010 165.2800 166.9100 179.9100 165.4600 173.8900 171.0000 162.5300 166.6500
## 2011 176.9600 179.4600 190.7100 175.1800 184.3000 182.3300 175.8300 185.6700
## 2012 181.5100 189.2500 202.5200 183.8100 193.4500 192.2800 185.8900 193.6100
## 2013 189.6800 192.6600 196.3700 195.4900 199.0000 194.3800 190.4500 196.6600
## 2014 194.2000 197.5800 205.4100 197.3600 207.0300 198.0900 194.1800 199.2100
## 2015 200.8200 202.0200 214.0600 206.3900 206.6600 206.1300 201.9400 207.7800
## 2016 207.8700 210.5600 220.5100 211.0700 214.4500 216.0000 205.6100 215.9800
## 2017 219.3700 221.5000 233.9300 218.0300 225.5300 225.9000 216.7500 229.0800
## 2018 228.9700 228.1200 237.1100 227.1200 234.8800 234.0300 225.0400 238.6600
## 2019 235.3000 235.0800 246.4000 234.8000 241.5100 235.4600 238.0200 244.6500
## 2020 242.4900 241.6500 218.2700 186.8800 189.0700 208.7100 209.3000 225.8000
## 2021 229.9700 236.2800 251.0500 235.9600 242.3600 247.4000 239.8100 256.7700
## 2022 247.2700 246.6200 263.7000 248.7700 254.7300 256.2300 246.2000 272.1400
## 2023 260.0353 260.4971
## Sep Oct Nov Dec
## 2009 157.8200 166.3300 163.9700 176.1600
## 2010 175.1800 172.0000 175.4800 186.8900
## 2011 182.0300 185.8200 188.1800 198.6600
## 2012 188.7900 199.9700 199.4800 203.1000
## 2013 191.3200 201.7900 201.5400 213.5700
## 2014 197.7300 205.5000 203.2600 221.7200
## 2015 204.9100 213.8100 214.7300 231.4000
## 2016 212.3100 220.7600 227.5900 245.5800
## 2017 226.2600 232.7500 235.8000 251.2300
## 2018 232.5500 244.9300 245.1600 262.4800
## 2019 239.6900 252.7200 250.2600 273.8000
## 2020 230.2400 249.3400 218.8900 258.0800
## 2021 246.8700 265.4500 264.7300 279.0500
## 2022 265.0962 277.6778 269.2452 289.4112
## 20234.4. Descomposición de la serie temporal
library(stats)
fit_hnd <- stl(ivae_h_hnd,"periodic")
autoplot(fit_hnd) + theme_bw()
TC_hnd <- fit_hnd$time.series[,2]
print(TC_hnd)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 162.8929 162.9561 163.0194 163.1303 163.2413 163.4432 163.6452 163.9242
## 2010 167.0260 168.0565 169.0871 170.0043 170.9216 171.6407 172.3599 173.0451
## 2011 177.8868 179.1969 180.5071 181.5959 182.6848 183.3904 184.0961 184.6435
## 2012 188.4348 189.4048 190.3749 191.3038 192.2328 192.7783 193.3238 193.4674
## 2013 194.8125 195.2254 195.6384 196.0381 196.4378 196.8346 197.2315 197.5486
## 2014 199.5783 200.0938 200.6092 201.0205 201.4318 201.7751 202.1184 202.4675
## 2015 205.0900 205.8980 206.7060 207.4727 208.2393 208.8819 209.5246 210.0341
## 2016 213.0173 213.7139 214.4105 215.1539 215.8972 216.7680 217.6388 218.5649
## 2017 223.3707 224.4099 225.4492 226.2883 227.1275 227.7553 228.3831 228.9430
## 2018 232.1604 232.9145 233.6685 234.4485 235.2286 236.0122 236.7959 237.4830
## 2019 240.4423 241.0225 241.6028 242.1963 242.7897 243.4264 244.0631 243.7318
## 2020 232.0460 229.5277 227.0094 225.2767 223.5440 222.7540 221.9639 223.2827
## 2021 237.1844 239.4830 241.7816 243.9924 246.2031 248.2298 250.2565 251.5898
## 2022 256.2256 257.0716 257.9175 258.9447 259.9720 261.0707 262.1694 263.1515
## 2023 268.2480 269.2274
## Sep Oct Nov Dec
## 2009 164.2033 164.7060 165.2087 166.1173
## 2010 173.7303 174.6313 175.5324 176.7096
## 2011 185.1909 185.9133 186.6357 187.5352
## 2012 193.6110 193.8109 194.0107 194.4116
## 2013 197.8658 198.2317 198.5976 199.0879
## 2014 202.8165 203.2806 203.7446 204.4173
## 2015 210.5437 211.0889 211.6342 212.3257
## 2016 219.4910 220.4128 221.3345 222.3526
## 2017 229.5029 230.1126 230.7224 231.4414
## 2018 238.1702 238.7462 239.3221 239.8822
## 2019 243.4005 240.9754 238.5503 235.2982
## 2020 224.6016 227.8072 231.0128 234.0986
## 2021 252.9232 253.8282 254.7332 255.4794
## 2022 264.1337 265.1653 266.1970 267.2225
## 20234.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_hnd %>% as.numeric() %>% as.data.frame() -> TC_df_hnd
names(TC_df_hnd) <- c("TC_hnd")
TC_df_hnd %>% mutate(T_1_1 = (TC_hnd/dplyr::lag(TC_hnd, n = 1) - 1) * 100,
T_1_12 = (TC_hnd/dplyr::lag(TC_hnd, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_hnd, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_hnd, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_hnd
tail(tabla_coyuntura_hnd, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_hnd | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 257.9175 | 0.3290682 | 6.673751 | 10.013832 | 4.432366 | NA |
| 258.9447 | 0.3982731 | 6.128209 | 9.810425 | 4.466462 | NA |
| 259.9720 | 0.3966932 | 5.592463 | 9.412695 | 4.500315 | NA |
| 261.0707 | 0.4226321 | 5.172975 | 8.879544 | 4.596477 | NA |
| 262.1694 | 0.4208535 | 4.760281 | 8.216186 | 4.692079 | NA |
| 263.1515 | 0.3746166 | 4.595455 | 7.553127 | 4.728556 | NA |
| 264.1337 | 0.3732185 | 4.432366 | 6.890368 | NA | NA |
| 265.1653 | 0.3905831 | 4.466462 | 6.328132 | NA | NA |
| 266.1970 | 0.3890635 | 4.500315 | 5.862449 | NA | NA |
| 267.2225 | 0.3852338 | 4.596477 | 5.496442 | NA | NA |
| 268.2480 | 0.3837555 | 4.692079 | 5.227119 | NA | NA |
| 269.2274 | 0.3651127 | 4.728556 | 5.016246 | NA | NA |
4.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_hnd %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_hnd
autoplot(tabla_coyuntura_graficos_hnd) + theme_bw()
tabla_coyuntura_hnd %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
5. Nicaragua
5.1. Crear data de Nicaragua
data_nic <- pivot_longer(data = serie_IMAE[,6],
cols = `Nicaragua`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_nic <- data_nic[1:164,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_nic)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 105.02 99.61 99.82 97.86 105.33 102.92 111.29 106.11 100.80 103.08
## 2010 107.22 102.04 106.17 100.25 108.47 107.98 116.44 110.70 106.35 110.07
## 2011 112.84 105.40 114.89 106.19 118.57 116.46 126.36 118.61 112.82 113.74
## 2012 128.39 116.85 118.64 112.51 126.31 118.10 130.29 123.88 117.08 126.20
## 2013 132.07 122.40 122.30 126.76 132.79 123.18 138.36 130.19 125.12 130.05
## 2014 135.68 129.80 132.03 128.86 139.04 130.03 143.73 133.05 131.23 137.49
## 2015 141.73 135.06 139.10 131.32 143.71 134.69 151.29 141.67 141.01 146.60
## 2016 148.01 141.73 143.00 140.87 153.13 144.24 155.81 149.66 143.57 149.07
## 2017 159.90 150.21 154.66 144.21 159.98 150.52 161.86 154.39 147.57 154.82
## 2018 165.61 154.20 158.41 150.62 151.56 130.54 153.23 148.98 141.06 143.07
## 2019 151.81 138.11 139.71 137.92 145.19 135.01 150.33 143.56 138.82 147.62
## 2020 153.26 145.11 140.70 124.93 134.77 130.15 148.61 139.39 140.78 148.10
## 2021 155.54 148.15 152.24 145.97 159.52 155.10 165.84 154.92 151.40 160.62
## 2022 166.47 154.51 161.07 153.46 166.82 160.07 171.29 161.96
## Nov Dec
## 2009 109.25 120.21
## 2010 116.56 124.67
## 2011 125.90 128.32
## 2012 130.71 142.11
## 2013 134.02 147.29
## 2014 141.38 157.08
## 2015 148.63 163.14
## 2016 155.85 171.41
## 2017 164.86 176.56
## 2018 153.82 165.28
## 2019 154.00 165.59
## 2020 145.86 164.76
## 2021 165.95 178.63
## 2022library(ggplot2)
autoplot(imae_nic,
xlab = "años",
ylab = "Indice",
main = "IMAE total de Nicaragua, periodo 2009 - 2022 (agosto)") +
theme_bw()
5.2. Proyección a seis meses
library(forecast)
modelo_nic <- auto.arima(y = imae_nic)
summary(modelo_nic)## Series: imae_nic
## ARIMA(0,1,1)(0,1,1)[12]
##
## Coefficients:
## ma1 sma1
## -0.3543 -0.7814
## s.e. 0.0847 0.0856
##
## sigma^2 = 13.91: log likelihood = -417.72
## AIC=841.44 AICc=841.6 BIC=850.49
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set -0.007934041 3.554758 2.510458 -0.04869239 1.813168 0.3469974
## ACF1
## Training set 0.02827215Se tiene:
\(ARIMA(0, 1, 1)(0, 1, 1)[12]\)
pronosticos_nic <- forecast(modelo_nic, h = 6)
autoplot(pronosticos_nic) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_nic$x, series = "IMAE Nicaragua") + autolayer(pronosticos_nic$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
5.3. Serie ampliada
ivae_h_nic <- ts(as.numeric(rbind(as.matrix(pronosticos_nic$x),
as.matrix(pronosticos_nic$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_nic)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 105.0200 99.6100 99.8200 97.8600 105.3300 102.9200 111.2900 106.1100
## 2010 107.2200 102.0400 106.1700 100.2500 108.4700 107.9800 116.4400 110.7000
## 2011 112.8400 105.4000 114.8900 106.1900 118.5700 116.4600 126.3600 118.6100
## 2012 128.3900 116.8500 118.6400 112.5100 126.3100 118.1000 130.2900 123.8800
## 2013 132.0700 122.4000 122.3000 126.7600 132.7900 123.1800 138.3600 130.1900
## 2014 135.6800 129.8000 132.0300 128.8600 139.0400 130.0300 143.7300 133.0500
## 2015 141.7300 135.0600 139.1000 131.3200 143.7100 134.6900 151.2900 141.6700
## 2016 148.0100 141.7300 143.0000 140.8700 153.1300 144.2400 155.8100 149.6600
## 2017 159.9000 150.2100 154.6600 144.2100 159.9800 150.5200 161.8600 154.3900
## 2018 165.6100 154.2000 158.4100 150.6200 151.5600 130.5400 153.2300 148.9800
## 2019 151.8100 138.1100 139.7100 137.9200 145.1900 135.0100 150.3300 143.5600
## 2020 153.2600 145.1100 140.7000 124.9300 134.7700 130.1500 148.6100 139.3900
## 2021 155.5400 148.1500 152.2400 145.9700 159.5200 155.1000 165.8400 154.9200
## 2022 166.4700 154.5100 161.0700 153.4600 166.8200 160.0700 171.2900 161.9600
## 2023 172.5742 162.9901
## Sep Oct Nov Dec
## 2009 100.8000 103.0800 109.2500 120.2100
## 2010 106.3500 110.0700 116.5600 124.6700
## 2011 112.8200 113.7400 125.9000 128.3200
## 2012 117.0800 126.2000 130.7100 142.1100
## 2013 125.1200 130.0500 134.0200 147.2900
## 2014 131.2300 137.4900 141.3800 157.0800
## 2015 141.0100 146.6000 148.6300 163.1400
## 2016 143.5700 149.0700 155.8500 171.4100
## 2017 147.5700 154.8200 164.8600 176.5600
## 2018 141.0600 143.0700 153.8200 165.2800
## 2019 138.8200 147.6200 154.0000 165.5900
## 2020 140.7800 148.1000 145.8600 164.7600
## 2021 151.4000 160.6200 165.9500 178.6300
## 2022 158.8882 165.6427 170.7768 184.1819
## 20235.4. Descomposición de la serie temporal
library(stats)
fit_nic <- stl(ivae_h_nic,"periodic")
autoplot(fit_nic) + theme_bw()
TC_nic <- fit_nic$time.series[,2]
print(TC_nic)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 103.7474 104.0257 104.3039 104.5511 104.7982 105.0392 105.2803 105.5336
## 2010 106.8872 107.3985 107.9098 108.5315 109.1533 109.6689 110.1845 110.5973
## 2011 113.3401 114.0861 114.8322 115.5596 116.2870 117.0083 117.7296 118.3527
## 2012 120.6025 121.1265 121.6505 122.3302 123.0099 123.6561 124.3023 124.8340
## 2013 127.7848 128.3827 128.9805 129.4640 129.9475 130.3456 130.7437 131.1448
## 2014 133.4107 133.9444 134.4781 135.0743 135.6705 136.2535 136.8365 137.2926
## 2015 139.5101 140.1693 140.8286 141.5379 142.2473 142.8757 143.5041 144.0480
## 2016 146.9604 147.4556 147.9509 148.4325 148.9141 149.5811 150.2482 150.9694
## 2017 154.0120 154.5054 154.9989 155.4324 155.8658 156.2886 156.7114 157.1097
## 2018 156.4340 155.5569 154.6798 153.6371 152.5944 151.4120 150.2296 149.0951
## 2019 145.8680 145.6309 145.3937 145.3315 145.2693 145.4599 145.6505 145.8659
## 2020 144.4457 143.9696 143.4935 143.1543 142.8152 142.7984 142.7815 143.5215
## 2021 150.7900 152.1981 153.6062 154.8379 156.0696 157.1453 158.2209 158.9661
## 2022 161.8980 162.4347 162.9715 163.4583 163.9451 164.4453 164.9456 165.4437
## 2023 167.8960 168.3823
## Sep Oct Nov Dec
## 2009 105.7869 106.0016 106.2163 106.5518
## 2010 111.0100 111.4965 111.9830 112.6615
## 2011 118.9758 119.4013 119.8267 120.2146
## 2012 125.3657 125.9593 126.5530 127.1689
## 2013 131.5459 131.9884 132.4309 132.9208
## 2014 137.7487 138.1401 138.5316 139.0208
## 2015 144.5919 145.1781 145.7643 146.3623
## 2016 151.6907 152.3040 152.9173 153.4646
## 2017 157.5081 157.5538 157.5996 157.0168
## 2018 147.9607 147.2269 146.4931 146.1806
## 2019 146.0813 145.8251 145.5689 145.0073
## 2020 144.2615 145.8039 147.3463 149.0681
## 2021 159.7112 160.2728 160.8344 161.3662
## 2022 165.9419 166.4320 166.9222 167.4091
## 20235.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_nic %>% as.numeric() %>% as.data.frame() -> TC_df_nic
names(TC_df_nic) <- c("TC_nic")
TC_df_nic %>% mutate(T_1_1 = (TC_nic/dplyr::lag(TC_nic, n = 1) - 1) * 100,
T_1_12 = (TC_nic/dplyr::lag(TC_nic, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_nic, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_nic, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_nic
tail(tabla_coyuntura_nic, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_nic | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 162.9715 | 0.3304316 | 6.096918 | 8.906175 | 3.901214 | NA |
| 163.4583 | 0.2987148 | 5.567363 | 8.674304 | 3.842958 | NA |
| 163.9451 | 0.2978251 | 5.046167 | 8.308167 | 3.785108 | NA |
| 164.4453 | 0.3051179 | 4.645430 | 7.849802 | 3.744830 | NA |
| 164.9456 | 0.3041898 | 4.250143 | 7.302228 | 3.704817 | NA |
| 165.4437 | 0.3020289 | 4.074861 | 6.750095 | 3.661519 | NA |
| 165.9419 | 0.3011195 | 3.901214 | 6.193522 | NA | NA |
| 166.4320 | 0.2953553 | 3.842958 | 5.698114 | NA | NA |
| 166.9222 | 0.2944855 | 3.785108 | 5.261674 | NA | NA |
| 167.4091 | 0.2917101 | 3.744830 | 4.895316 | NA | NA |
| 167.8960 | 0.2908616 | 3.704817 | 4.596985 | NA | NA |
| 168.3823 | 0.2896379 | 3.661519 | 4.347184 | NA | NA |
5.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_nic %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_nic
autoplot(tabla_coyuntura_graficos_nic) + theme_bw()
tabla_coyuntura_nic %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
6. Panamá
6.1. Crear data de Panamá
data_pan <- pivot_longer(data = serie_IMAE[,7],
cols = `Panamá`,
names_to = "pais",
values_to = "indice") %>% select("indice")
imae_pan <- data_pan[1:164,] %>% ts(start = c(2009,1),frequency = 12)
print(imae_pan)## Jan Feb Mar Apr May Jun Jul Aug Sep Oct
## 2009 195.71 189.71 204.64 188.06 193.65 199.60 188.20 190.96 195.45 204.89
## 2010 201.01 200.42 220.03 203.11 202.73 210.75 198.31 206.22 205.20 213.91
## 2011 212.36 218.50 228.61 218.93 227.12 226.92 210.41 225.69 222.92 233.74
## 2012 233.23 237.88 260.05 237.89 248.64 251.02 239.86 246.47 238.38 249.62
## 2013 253.29 254.04 276.60 262.60 268.38 269.16 256.27 265.08 259.72 280.51
## 2014 265.09 267.31 286.56 275.53 274.45 283.35 268.30 278.43 272.53 296.66
## 2015 281.48 276.75 307.31 280.85 281.06 294.76 279.85 290.40 283.40 310.57
## 2016 292.53 289.67 318.79 292.13 296.56 306.77 293.76 303.34 296.96 322.82
## 2017 305.84 307.27 344.01 309.60 316.06 324.68 304.97 318.19 310.13 335.94
## 2018 320.57 323.91 349.99 311.69 317.94 324.94 308.98 323.54 315.15 333.20
## 2019 332.39 332.73 353.65 319.17 325.72 332.45 325.11 336.06 332.01 346.53
## 2020 346.02 341.78 357.06 243.21 222.17 233.12 240.66 242.56 259.86 298.72
## 2021 304.59 322.77 354.90 307.26 314.67 309.91 306.48 318.85 317.91 344.08
## 2022 354.01 368.38 390.50 334.96 344.38 348.03 317.45 359.51
## Nov Dec
## 2009 185.82 190.56
## 2010 202.49 205.63
## 2011 226.59 231.09
## 2012 251.28 247.71
## 2013 272.24 270.52
## 2014 282.62 292.03
## 2015 295.47 300.89
## 2016 309.18 312.22
## 2017 322.13 324.80
## 2018 328.79 330.41
## 2019 341.09 341.27
## 2020 296.58 339.78
## 2021 332.46 395.90
## 2022library(ggplot2)
autoplot(imae_pan,
xlab = "años",
ylab = "Indice",
main = "IMAE total de Panamá, periodo 2009 - 2022 (agosto)") +
theme_bw()
6.2. Proyección a seis meses
library(forecast)
modelo_pan <- auto.arima(y = imae_pan)
summary(modelo_pan)## Series: imae_pan
## ARIMA(1,0,2)(0,1,1)[12] with drift
##
## Coefficients:
## ar1 ma1 ma2 sma1 drift
## 0.8591 -0.0045 0.1297 -0.4593 1.0167
## s.e. 0.0537 0.1011 0.0880 0.1003 0.3442
##
## sigma^2 = 135.4: log likelihood = -588.3
## AIC=1188.6 AICc=1189.18 BIC=1206.75
##
## Training set error measures:
## ME RMSE MAE MPE MAPE MASE
## Training set 0.07830014 11.0181 5.999723 -0.06832539 2.122395 0.2800424
## ACF1
## Training set 0.000449099Se tiene:
\(ARIMA(1, 0, 2)(0, 1, 1)[12]\)
pronosticos_pan <- forecast(modelo_pan, h = 6)
autoplot(pronosticos_pan) + xlab("Años") + ylab("indice") + theme_bw()
library(forecast)
autoplot(pronosticos_pan$x, series = "IMAE Panamá") + autolayer(pronosticos_pan$fitted, series = "Pronóstico") + ggtitle("Ajuste SARIMA")
6.3. Serie ampliada
ivae_h_pan <- ts(as.numeric(rbind(as.matrix(pronosticos_pan$x),
as.matrix(pronosticos_pan$mean))),
start = c(2009,1),
frequency = 12)
print(ivae_h_pan)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 195.7100 189.7100 204.6400 188.0600 193.6500 199.6000 188.2000 190.9600
## 2010 201.0100 200.4200 220.0300 203.1100 202.7300 210.7500 198.3100 206.2200
## 2011 212.3600 218.5000 228.6100 218.9300 227.1200 226.9200 210.4100 225.6900
## 2012 233.2300 237.8800 260.0500 237.8900 248.6400 251.0200 239.8600 246.4700
## 2013 253.2900 254.0400 276.6000 262.6000 268.3800 269.1600 256.2700 265.0800
## 2014 265.0900 267.3100 286.5600 275.5300 274.4500 283.3500 268.3000 278.4300
## 2015 281.4800 276.7500 307.3100 280.8500 281.0600 294.7600 279.8500 290.4000
## 2016 292.5300 289.6700 318.7900 292.1300 296.5600 306.7700 293.7600 303.3400
## 2017 305.8400 307.2700 344.0100 309.6000 316.0600 324.6800 304.9700 318.1900
## 2018 320.5700 323.9100 349.9900 311.6900 317.9400 324.9400 308.9800 323.5400
## 2019 332.3900 332.7300 353.6500 319.1700 325.7200 332.4500 325.1100 336.0600
## 2020 346.0200 341.7800 357.0600 243.2100 222.1700 233.1200 240.6600 242.5600
## 2021 304.5900 322.7700 354.9000 307.2600 314.6700 309.9100 306.4800 318.8500
## 2022 354.0100 368.3800 390.5000 334.9600 344.3800 348.0300 317.4500 359.5100
## 2023 377.5661 387.0177
## Sep Oct Nov Dec
## 2009 195.4500 204.8900 185.8200 190.5600
## 2010 205.2000 213.9100 202.4900 205.6300
## 2011 222.9200 233.7400 226.5900 231.0900
## 2012 238.3800 249.6200 251.2800 247.7100
## 2013 259.7200 280.5100 272.2400 270.5200
## 2014 272.5300 296.6600 282.6200 292.0300
## 2015 283.4000 310.5700 295.4700 300.8900
## 2016 296.9600 322.8200 309.1800 312.2200
## 2017 310.1300 335.9400 322.1300 324.8000
## 2018 315.1500 333.2000 328.7900 330.4100
## 2019 332.0100 346.5300 341.0900 341.2700
## 2020 259.8600 298.7200 296.5800 339.7800
## 2021 317.9100 344.0800 332.4600 395.9000
## 2022 353.6559 381.5419 369.4203 411.3908
## 20236.4. Descomposición de la serie temporal
library(stats)
fit_pan <- stl(ivae_h_pan,"periodic")
autoplot(fit_pan) + theme_bw()
TC_pan <- fit_pan$time.series[,2]
print(TC_pan)## Jan Feb Mar Apr May Jun Jul Aug
## 2009 192.5034 193.0118 193.5203 193.9132 194.3061 194.6900 195.0739 195.5994
## 2010 199.1110 200.4670 201.8231 203.2822 204.7414 205.9343 207.1272 207.8752
## 2011 213.2980 215.1717 217.0455 219.1430 221.2405 223.1781 225.1157 226.6507
## 2012 235.3317 237.5384 239.7452 241.7110 243.6768 245.2283 246.7798 247.8806
## 2013 254.6459 256.7700 258.8942 261.1832 263.4723 265.1834 266.8945 267.6906
## 2014 271.1301 272.4223 273.7144 275.2501 276.7857 278.3031 279.8204 280.8491
## 2015 284.3681 285.4116 286.4550 287.7133 288.9715 290.1608 291.3500 292.2442
## 2016 296.4131 297.5917 298.7703 300.1445 301.5187 302.8779 304.2371 305.5342
## 2017 312.0120 313.3711 314.7301 316.1266 317.5231 318.7218 319.9204 320.6350
## 2018 322.1079 322.4591 322.8103 323.2576 323.7049 324.2519 324.7989 325.3460
## 2019 328.5264 329.7135 330.9007 332.1880 333.4753 334.5676 335.6599 335.6851
## 2020 313.0948 304.9626 296.8305 290.7894 284.7484 281.6624 278.5763 279.4113
## 2021 303.7174 308.8374 313.9574 317.6293 321.3012 324.7792 328.2573 332.0488
## 2022 348.3066 350.6275 352.9484 354.9738 356.9993 359.6712 362.3430 364.9106
## 2023 379.0465 382.2866
## Sep Oct Nov Dec
## 2009 196.1249 196.7355 197.3461 198.2286
## 2010 208.6231 209.5807 210.5384 211.9182
## 2011 228.1857 229.7888 231.3920 233.3618
## 2012 248.9814 250.1836 251.3857 253.0158
## 2013 268.4868 268.9704 269.4540 270.2921
## 2014 281.8779 282.4382 282.9985 283.6833
## 2015 293.1385 293.8552 294.5720 295.4925
## 2016 306.8314 308.0977 309.3641 310.6881
## 2017 321.3495 321.5466 321.7437 321.9258
## 2018 325.8932 326.4139 326.9346 327.7305
## 2019 335.7102 331.8942 328.0782 320.5865
## 2020 280.2463 285.3590 290.4716 297.0945
## 2021 335.8404 339.2806 342.7208 345.5137
## 2022 367.4782 370.2250 372.9719 376.0092
## 20236.5. Cálculo de las tasas
library(dplyr)
library(zoo)
TC_pan %>% as.numeric() %>% as.data.frame() -> TC_df_pan
names(TC_df_pan) <- c("TC_pan")
TC_df_pan %>% mutate(T_1_1 = (TC_pan/dplyr::lag(TC_pan, n = 1) - 1) * 100,
T_1_12 = (TC_pan/dplyr::lag(TC_pan, n = 12) - 1) * 100,
T_12_12 = (rollapply(TC_pan, 12, mean,
align = 'right', fill = NA)
/rollapply(dplyr::lag(TC_pan, n = 12), 12, mean,
align = 'right', fill = NA) - 1) * 100) %>%
#Aquí 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(2009, 1), frequency = 12) -> tabla_coyuntura_pan
tail(tabla_coyuntura_pan, n = 12) %>% kable(align = "l") %>%
kable_material(html_font = "Times New Roman") %>% kable_styling()| TC_pan | T_1_1 | T_1_12 | T_12_12 | T_1_12C | T_12_12C |
|---|---|---|---|---|---|
| 352.9484 | 0.6619302 | 12.419185 | 15.576603 | 9.420485 | NA |
| 354.9738 | 0.5738609 | 11.757267 | 15.756172 | 9.120609 | NA |
| 356.9993 | 0.5705865 | 11.110477 | 15.570294 | 8.826752 | NA |
| 359.6712 | 0.7484281 | 10.743272 | 15.155499 | 8.826136 | NA |
| 362.3430 | 0.7428683 | 10.383848 | 14.522132 | 8.825529 | NA |
| 364.9106 | 0.7086040 | 9.896666 | 13.781796 | 9.029259 | NA |
| 367.4782 | 0.7036182 | 9.420485 | 12.940730 | NA | NA |
| 370.2250 | 0.7474859 | 9.120609 | 12.155172 | NA | NA |
| 372.9719 | 0.7419400 | 8.826752 | 11.421780 | NA | NA |
| 376.0092 | 0.8143509 | 8.826136 | 10.822528 | NA | NA |
| 379.0465 | 0.8077728 | 8.825529 | 10.350756 | NA | NA |
| 382.2866 | 0.8547957 | 9.029259 | 9.989364 | NA | NA |
6.6. Gráfico de las tasas (centradas)
library(dplyr)
library(forecast)
library(ggplot2)
tabla_coyuntura_pan %>% as.data.frame() %>%
select(T_1_12C, T_12_12C) %>%
ts(start = c(2009,1), frequency = 12) -> tabla_coyuntura_graficos_pan
autoplot(tabla_coyuntura_graficos_pan) + theme_bw()
tabla_coyuntura_pan %>% as.data.frame() %>% select(T_1_1) %>% ts(start = c(2009,1),frequency = 12) %>% autoplot()
Gráfico de todas las tasas (centradas) - comparativa
tabla_coyuntura_graficos_sv -> tcg_SV
tabla_coyuntura_graficos_cr -> tcg_CR
tabla_coyuntura_graficos_gtm -> tcg_GTM
tabla_coyuntura_graficos_hnd -> tcg_HND
tabla_coyuntura_graficos_nic -> tcg_NIC
tabla_coyuntura_graficos_pan -> tcg_PAN
tabla_coyuntura_graficos_comparativa <- cbind(tcg_SV,
tcg_CR,
tcg_GTM,
tcg_HND,
tcg_NIC,
tcg_PAN)
autoplot(tabla_coyuntura_graficos_comparativa) + theme_bw()
tabla_coyuntura_sv %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_SV
tabla_coyuntura_cr %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_CR
tabla_coyuntura_gtm %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_GTM
tabla_coyuntura_hnd %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_HND
tabla_coyuntura_nic %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_NIC
tabla_coyuntura_pan %>% as.data.frame() %>% select(T_1_12) %>% ts(start = c(2009,1),frequency = 12) -> tc_PAN
tabla_coyuntura_grafico_comparativa_T_1_2 <- cbind(tc_SV,
tc_CR,
tc_GTM,
tc_HND,
tc_NIC,
tc_PAN)
autoplot(tabla_coyuntura_grafico_comparativa_T_1_2)