Carga de Datos
IVAE_03_22 <-
read_excel(
"C:/Users/Keiry/Documents/Eco22/IVAE_22_03.xlsx",
col_names = FALSE,
skip = 6,
n_max = 10
)
IVAE_03_22[1:10, 1:10]
## # A tibble: 10 x 10
## ...1 ...2 ...3 ...4 ...5 ...6 ...7 ...8 ...9 ...10
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 1 IVAE 86.7 80.8 87.2 83.9 91.4 93.5 86.4 86.7 87.6
## 2 2 Agricultura, Gan~ 86.8 70.2 65.2 67.4 138. 153. 75.6 137. 110.
## 3 3 Índice de Produc~ 86.6 85.4 100. 91.6 91.2 89.7 86.8 81.2 86.6
## 4 4 Construcción 66.6 76.9 82.6 82.6 69.0 80.8 82.5 67.3 74.5
## 5 5 Comercio, Transp~ 91.0 74.8 77.8 80.0 89.6 90.6 81.9 82.3 80.1
## 6 6 Información y Co~ 94.6 76.9 81.9 81.0 92.3 91.9 96.9 81.3 104.
## 7 7 Actividades Fina~ 97.4 85.7 94.9 88.9 92.6 91.6 92.1 94.2 92.6
## 8 8 Actividades Inmo~ 96.8 95.2 94.6 93.9 94.0 94.6 95.3 95.9 96.4
## 9 9 Actividades Prof~ 80.7 76.9 81.0 77.2 86.1 87.8 82.9 75.9 80.4
## 10 10 Actividades de ~ 80.8 83.0 91.3 83.9 84.9 86.7 90.3 87.7 88.4
Prónostico Modelo de Holt Winters “multiplicative”
Usando Stars y forescaste
#Carga de datos
data.ivae <-
pivot_longer(
data = IVAE_03_22[1,],
names_to = "vars",
cols = 2:160,
values_to = "indice"
) %>% select("indice")
data.ivae.ts <- data.ivae %>% ts(start = c(2009, 1), frequency = 12)
#Estimacion del modelo
ModeloHW <-
HoltWinters(x = data.ivae.ts,
seasonal = "multiplicative",
optim.start = c(0.9, 0.9, 0.9))
ModeloHW
## Holt-Winters exponential smoothing with trend and multiplicative seasonal component.
##
## Call:
## HoltWinters(x = data.ivae.ts, seasonal = "multiplicative", optim.start = c(0.9, 0.9, 0.9))
##
## Smoothing parameters:
## alpha: 0.8472467
## beta : 0
## gamma: 1
##
## Coefficients:
## [,1]
## a 118.3719784
## b 0.1600947
## s1 0.9529095
## s2 1.0192349
## s3 1.0440936
## s4 0.9962326
## s5 1.0009929
## s6 0.9838373
## s7 0.9554392
## s8 1.0145286
## s9 1.0872315
## s10 0.9748891
## s11 0.9626701
## s12 1.0059813
#Generar el pronostico
PronosticoHW <-
forecast(object = ModeloHW,
h = 12,
level = c(0.95, 0.99))
PronosticoHW
## Point Forecast Lo 95 Hi 95 Lo 99 Hi 99
## Apr 2022 112.9503 107.46492 118.4358 105.74127 120.1594
## May 2022 120.9752 113.57246 128.3779 111.24635 130.7040
## Jun 2022 124.0929 115.22236 132.9634 112.43504 135.7507
## Jul 2022 118.5640 108.86876 128.2592 105.82229 131.3057
## Aug 2022 119.2908 108.50116 130.0804 105.11081 133.4708
## Sep 2022 117.4038 105.81297 128.9947 102.17086 132.6368
## Oct 2022 114.1680 101.97016 126.3657 98.13733 130.1986
## Nov 2022 121.3911 107.66977 135.1125 103.35822 139.4240
## Dec 2022 130.2643 114.88360 145.6450 110.05064 150.4779
## Jan 2023 116.9603 102.34981 131.5708 97.75887 136.1617
## Feb 2023 115.6485 100.48405 130.8129 95.71905 135.5779
## Mar 2023 121.0126 83.19827 158.8270 71.31613 170.7091
#Grafico de la serie orifinal y del pronostico
PronosticoHW %>% autoplot()
Usando Forescast (Aprocimacion por Espacios por los Estados ETS)
#Generar el pronostico
PronosticoHW2 <-
hw(
y = data.ivae.ts,
h = 12,
level = c(0.96, 0.99),
seasonal = "multiplicative",
initial = "optimal"
)
PronosticoHW2
## Point Forecast Lo 96 Hi 96 Lo 99 Hi 99
## Apr 2022 111.9184 104.76599 119.0709 102.94778 120.8891
## May 2022 119.3664 110.25864 128.4742 107.94336 130.7895
## Jun 2022 122.8422 112.14980 133.5347 109.43169 136.2528
## Jul 2022 118.4498 106.99770 129.9019 104.08648 132.8131
## Aug 2022 120.0189 107.35373 132.6840 104.13415 135.9036
## Sep 2022 118.4260 104.95412 131.8980 101.52944 135.3226
## Oct 2022 114.3087 100.41978 128.1975 96.88911 131.7282
## Nov 2022 120.2837 104.78516 135.7822 100.84531 139.7221
## Dec 2022 127.9776 110.59023 145.3649 106.17022 149.7849
## Jan 2023 114.4558 98.13547 130.7762 93.98669 134.9249
## Feb 2023 113.5923 96.65871 130.5259 92.35405 134.8306
## Mar 2023 119.3883 100.84228 137.9342 96.12773 142.6488
#Gráfico de la serie original y del pronóstico.
PronosticoHW2 %>% autoplot()
Prónostico Modelo de Holt Winters “additive”
Usando Stars y forescaste
#Carga de datos
data.ivae <-
pivot_longer(
data = IVAE_03_22[1,],
names_to = "vars",
cols = 2:160,
values_to = "indice"
) %>% select("indice")
data.ivae.ts <- data.ivae %>% ts(start = c(2009, 1), frequency = 12)
#Estimacion del modelo
ModeloHW <-
HoltWinters(x = data.ivae.ts,
seasonal = "additive",
optim.start = c(0.9, 0.9, 0.9))
ModeloHW
## Holt-Winters exponential smoothing with trend and additive seasonal component.
##
## Call:
## HoltWinters(x = data.ivae.ts, seasonal = "additive", optim.start = c(0.9, 0.9, 0.9))
##
## Smoothing parameters:
## alpha: 0.9142733
## beta : 0
## gamma: 1
##
## Coefficients:
## [,1]
## a 118.0938359
## b 0.1600947
## s1 -3.9032233
## s2 3.0467916
## s3 3.9273013
## s4 -1.6040517
## s5 -0.5787492
## s6 -1.8288955
## s7 -4.0808499
## s8 2.2490593
## s9 9.6275698
## s10 -2.7962367
## s11 -4.4557494
## s12 0.9861641
#Generar el pronostico
PronosticoHW <-
forecast(object = ModeloHW,
h = 12,
level = c(0.95, 0.99))
PronosticoHW
## Point Forecast Lo 95 Hi 95 Lo 99 Hi 99
## Apr 2022 114.3507 109.10096 119.6005 107.45137 121.2500
## May 2022 121.4608 114.34766 128.5740 112.11255 130.8091
## Jun 2022 122.5014 113.92039 131.0825 111.22404 133.7788
## Jul 2022 117.1302 107.29801 126.9623 104.20853 130.0518
## Aug 2022 118.3156 107.37443 129.2567 103.93648 132.6946
## Sep 2022 117.2255 105.27790 129.1731 101.52368 132.9273
## Oct 2022 115.1336 102.25799 128.0093 98.21216 132.0551
## Nov 2022 121.6237 107.88248 135.3648 103.56470 139.6826
## Dec 2022 129.1623 114.60695 143.7176 110.03335 148.2912
## Jan 2023 116.8985 101.57229 132.2248 96.75644 137.0407
## Feb 2023 115.3991 99.33889 131.4594 94.29240 136.5059
## Mar 2023 121.0011 104.23903 137.7632 98.97199 143.0303
#Grafico de la serie orifinal y del pronostico
PronosticoHW %>% autoplot()
Usando Forescast (Aprocimacion por Espacios por los Estados ETS)
#Generar el pronostico
PronosticoHW2 <-
hw(
y = data.ivae.ts,
h = 12,
level = c(0.96, 0.99),
seasonal = "additive",
initial = "optimal"
)
PronosticoHW2
## Point Forecast Lo 96 Hi 96 Lo 99 Hi 99
## Apr 2022 114.4467 109.34400 119.5493 108.04686 120.8465
## May 2022 120.9088 113.69268 128.1250 111.85827 129.9594
## Jun 2022 121.6963 112.85813 130.5345 110.61138 132.7813
## Jul 2022 116.4405 106.23463 126.6464 103.64021 129.2408
## Aug 2022 119.0528 107.64178 130.4638 104.74100 133.3646
## Sep 2022 118.3286 105.82785 130.8293 102.65006 134.0071
## Oct 2022 116.5899 103.08693 130.0929 99.65435 133.5255
## Nov 2022 122.4772 108.04118 136.9132 104.37143 140.5829
## Dec 2022 129.5220 114.20956 144.8344 110.31701 148.7270
## Jan 2023 117.7994 101.65791 133.9409 97.55460 138.0443
## Feb 2023 115.4411 98.51088 132.3713 94.20707 136.6751
## Mar 2023 121.4513 103.76743 139.1352 99.27204 143.6306
#Gráfico de la serie original y del pronóstico.
PronosticoHW2 %>% autoplot()
