Pronóstico de series temporales, enfoque determinístico (clásico)
JACQUELINE VANESSA CLAROS RUIZ
Serie (Yt)
library(readxl)
library(tidyr)
library(dplyr)
library(forecast)
IVAE_03_22 <- read_excel("IVAE_03_22.xlsx",
col_names = FALSE, skip = 6, n_max = 10)
data.ivae<-pivot_longer(data = IVAE_03_22[1,],names_to = "vars",cols = 2:160,values_to = "indice") %>% select("indice")
serie.ivae.ts<- data.ivae %>% ts(start = c(2009,1),frequency = 12)
Yt <- serie.ivae.ts
serie.ivae.ts %>%
autoplot(main ="IVAE ENE 2009- MAR 2022",
xlab="Años/Meses",
ylab="Indice")
Pronóstico de series temporales, enfoque determinístico (clásico)
Pronóstico Modelo de Holt Winters
Usando Stats y forecast
Multiplicativo
library(forecast)
#Estimar el modelo
ModeloHW_M<-HoltWinters(x = Yt,
seasonal = "multiplicative",
optim.start = c(0.9,0.9,0.9))
ModeloHW_M## Holt-Winters exponential smoothing with trend and multiplicative seasonal component.
##
## Call:
## HoltWinters(x = Yt, 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 pronóstico:
PronosticosHW_M<-forecast(object = ModeloHW_M,h = 12,level = c(0.95))
PronosticosHW_M## Point Forecast Lo 95 Hi 95
## Apr 2022 112.9503 107.46492 118.4358
## May 2022 120.9752 113.57246 128.3779
## Jun 2022 124.0929 115.22236 132.9634
## Jul 2022 118.5640 108.86876 128.2592
## Aug 2022 119.2908 108.50116 130.0804
## Sep 2022 117.4038 105.81297 128.9947
## Oct 2022 114.1680 101.97016 126.3657
## Nov 2022 121.3911 107.66977 135.1125
## Dec 2022 130.2643 114.88360 145.6450
## Jan 2023 116.9603 102.34981 131.5708
## Feb 2023 115.6485 100.48405 130.8129
## Mar 2023 121.0126 83.19827 158.8270#Gráfico de la serie original y del pronóstico.
PronosticosHW_M %>%
autoplot(main ="Pronóstico Modelo de Holt Winters - Multiplicativo",
xlab="Años/Meses",
ylab="Indice")
Aditivo
library(forecast)
#Estimar el modelo
ModeloHW_A<-HoltWinters(x = Yt,
seasonal = "additive",
optim.start = c(0.9,0.9,0.9))
ModeloHW_A## Holt-Winters exponential smoothing with trend and additive seasonal component.
##
## Call:
## HoltWinters(x = Yt, 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 pronóstico:
PronosticosHW_A<-forecast(object = ModeloHW_A,h = 12,level = c(0.95))
PronosticosHW_A## Point Forecast Lo 95 Hi 95
## Apr 2022 114.3507 109.10096 119.6005
## May 2022 121.4608 114.34766 128.5740
## Jun 2022 122.5014 113.92039 131.0825
## Jul 2022 117.1302 107.29801 126.9623
## Aug 2022 118.3156 107.37443 129.2567
## Sep 2022 117.2255 105.27790 129.1731
## Oct 2022 115.1336 102.25799 128.0093
## Nov 2022 121.6237 107.88248 135.3648
## Dec 2022 129.1623 114.60695 143.7176
## Jan 2023 116.8985 101.57229 132.2248
## Feb 2023 115.3991 99.33889 131.4594
## Mar 2023 121.0011 104.23903 137.7632#Gráfico de la serie original y del pronóstico.
PronosticosHW_A %>%
autoplot(main ="Pronóstico Modelo de Holt Winters - Aditivo",
xlab="Años/Meses",
ylab="Indice")
Usando Forecast (Aproximación por Espacios de los Estados ETS)
Multiplicativo
library(forecast)
#Generar el pronóstico:
PronosticosHW2_M<-hw(y = Yt,
h = 12,
level = c(0.95),
seasonal = "multiplicative",
initial = "optimal")
PronosticosHW2_M## Point Forecast Lo 95 Hi 95
## Apr 2022 111.9184 105.09260 118.7442
## May 2022 119.3664 110.67455 128.0583
## Jun 2022 122.8422 112.63807 133.0464
## Jul 2022 118.4498 107.52066 129.3789
## Aug 2022 120.0189 107.93209 132.1056
## Sep 2022 118.4260 105.56932 131.2828
## Oct 2022 114.3087 101.05402 127.5633
## Nov 2022 120.2837 105.49291 135.0745
## Dec 2022 127.9776 111.38423 144.5709
## Jan 2023 114.4558 98.88074 130.0309
## Feb 2023 113.5923 97.43199 129.7527
## Mar 2023 119.3883 101.68918 137.0873#Gráfico de la serie original y del pronóstico.
PronosticosHW2_M %>%
autoplot(main ="Pronóstico Modelo de Holt Winters - Multiplicativo",
xlab="Años/Meses",
ylab="Indice")
Aditivo
library(forecast)
#Generar el pronóstico:
PronosticosHW2_A<-hw(y = Yt,
h = 12,
level = c(0.95),
seasonal = "additive",
initial = "optimal")
PronosticosHW2_A## Point Forecast Lo 95 Hi 95
## Apr 2022 114.4467 109.5770 119.3163
## May 2022 120.9088 114.0222 127.7955
## Jun 2022 121.6963 113.2617 130.1309
## Jul 2022 116.4405 106.7007 126.1803
## Aug 2022 119.0528 108.1629 129.9428
## Sep 2022 118.3286 106.3987 130.2585
## Oct 2022 116.5899 103.7035 129.4763
## Nov 2022 122.4772 108.7004 136.2539
## Dec 2022 129.5220 114.9088 144.1352
## Jan 2023 117.7994 102.3950 133.2038
## Feb 2023 115.4411 99.2840 131.5982
## Mar 2023 121.4513 104.5750 138.3277#Gráfico de la serie original y del pronóstico.
PronosticosHW2_A %>%
autoplot(main ="Pronóstico Modelo de Holt Winters - Aditivo",
xlab="Años/Meses",
ylab="Indice")