1. Load example data from url:
ts <- read.csv("https://raw.githubusercontent.com/forvis/forvision_data/master/example1_TSTS.csv")
head(ts, 10)
## series_id value timestamp
## 1 Y1 3103.96 1984
## 2 Y1 3360.27 1985
## 3 Y1 3807.63 1986
## 4 Y1 4387.88 1987
## 5 Y1 4936.99 1988
## 6 Y1 5379.75 1989
## 7 Y1 6158.68 1990
## 8 Y1 6876.58 1991
## 9 Y2 5389.80 1984
## 10 Y2 5384.40 1985
fs <- read.csv("https://raw.githubusercontent.com/forvis/forvision_data/master/example1_FTS.csv")
head(fs, 18)
## series_id method timestamp origin_timestamp forecast horizon lo90
## 1 Y1 A 1989 1988 5406.43 1 5183.349
## 2 Y1 A 1990 1988 5875.96 2 5652.879
## 3 Y1 A 1991 1988 6345.48 3 6122.399
## 4 Y1 A 1990 1989 5946.48 1 6122.399
## 5 Y1 A 1991 1989 6145.48 2 6122.399
## 6 Y1 A 1991 1990 6445.48 1 6122.399
## 7 Y1 B 1989 1988 5473.87 1 5250.789
## 8 Y1 B 1990 1988 6010.43 2 5787.349
## 9 Y1 B 1991 1988 6546.63 3 6323.549
## 10 Y1 B 1990 1989 5373.87 1 5250.789
## 11 Y1 B 1991 1989 6113.87 2 5250.789
## 12 Y1 B 1991 1990 6473.87 1 5250.789
## 13 Y1 C 1989 1988 5406.43 1 5183.349
## 14 Y1 C 1990 1988 5875.96 2 5652.879
## 15 Y1 C 1991 1988 6345.48 3 6122.399
## 16 Y1 C 1990 1989 5806.43 1 5183.349
## 17 Y1 C 1991 1989 6106.43 2 5183.349
## 18 Y1 C 1991 1990 6406.43 1 5183.349
## hi90
## 1 5629.511
## 2 6099.041
## 3 6568.561
## 4 6568.561
## 5 6568.561
## 6 6568.561
## 7 5696.951
## 8 6233.511
## 9 6769.711
## 10 5696.951
## 11 5696.951
## 12 5696.951
## 13 5629.511
## 14 6099.041
## 15 6568.561
## 16 5629.511
## 17 5629.511
## 18 5629.511
2. Добавление и преобразование колонки timestamp_dbo в виде date-based object:
library(zoo)
ts$timestamp_dbo <- as.yearmon(ts$timestamp, format = '%Y')
fs$timestamp_dbo <- as.yearmon(fs$timestamp, '%Y')
head(ts)
## series_id value timestamp timestamp_dbo
## 1 Y1 3103.96 1984 Jan 1984
## 2 Y1 3360.27 1985 Jan 1985
## 3 Y1 3807.63 1986 Jan 1986
## 4 Y1 4387.88 1987 Jan 1987
## 5 Y1 4936.99 1988 Jan 1988
## 6 Y1 5379.75 1989 Jan 1989
head(fs)
## series_id method timestamp origin_timestamp forecast horizon lo90
## 1 Y1 A 1989 1988 5406.43 1 5183.349
## 2 Y1 A 1990 1988 5875.96 2 5652.879
## 3 Y1 A 1991 1988 6345.48 3 6122.399
## 4 Y1 A 1990 1989 5946.48 1 6122.399
## 5 Y1 A 1991 1989 6145.48 2 6122.399
## 6 Y1 A 1991 1990 6445.48 1 6122.399
## hi90 timestamp_dbo
## 1 5629.511 Jan 1989
## 2 6099.041 Jan 1990
## 3 6568.561 Jan 1991
## 4 6568.561 Jan 1990
## 5 6568.561 Jan 1991
## 6 6568.561 Jan 1991
3. plotFixedHorizon():
- не задаются аргументы
fs - FTS schema, horizon - forecast horizon, and m-forecast method то функция plotFixedHorizon() строит график выбранного временного ряда
library(forvision)
plotFixedHorizon(ts = ts, id ="Y1")
- задаются аргументы
fs - FTS schema, horizon - forecast hỏrizon and m-forecast method:
library(forvision)
plotFixedHorizon(ts = ts, fs = fs, id ="Y1", h = 1, m = "A")
library(forvision)
plotFixedHorizon(ts = ts, fs = fs, id ="Y1", h = 1, m = c("A", "C"))
library(forvision)
plotFixedHorizon(ts = ts, fs = fs, id ="Y1", h = 2, m = c("A", "B"))
4 Дополнить этот пример таблицами значений, по которым построены графики:
- сформировать переменную af=createAFTS(ts,fs)
ts <- read.csv("https://raw.githubusercontent.com/forvis/forvision_data/master/example1_TSTS.csv")
fs <- read.csv("https://raw.githubusercontent.com/forvis/forvision_data/master/example1_FTS.csv")
library(forvision)
af <- createAFTS(ts, fs)
head(af)
## series_id value timestamp method origin_timestamp forecast horizon
## 1 Y1 3103.96 1984 <NA> NA NA NA
## 2 Y1 3360.27 1985 <NA> NA NA NA
## 3 Y1 3807.63 1986 <NA> NA NA NA
## 4 Y1 4387.88 1987 <NA> NA NA NA
## 5 Y1 4936.99 1988 <NA> NA NA NA
## 6 Y1 5379.75 1989 A 1988 5406.43 1
## lo90 hi90
## 1 NA NA
## 2 NA NA
## 3 NA NA
## 4 NA NA
## 5 NA NA
## 6 5183.349 5629.511
subset(af,series_id=="Y1" & horizon==1)
## series_id value timestamp method origin_timestamp forecast horizon
## 6 Y1 5379.75 1989 A 1988 5406.43 1
## 7 Y1 5379.75 1989 B 1988 5473.87 1
## 8 Y1 5379.75 1989 C 1988 5406.43 1
## 10 Y1 6158.68 1990 A 1989 5946.48 1
## 12 Y1 6158.68 1990 B 1989 5373.87 1
## 14 Y1 6158.68 1990 C 1989 5806.43 1
## 17 Y1 6876.58 1991 A 1990 6445.48 1
## 20 Y1 6876.58 1991 B 1990 6473.87 1
## 23 Y1 6876.58 1991 C 1990 6406.43 1
## lo90 hi90
## 6 5183.349 5629.511
## 7 5250.789 5696.951
## 8 5183.349 5629.511
## 10 6122.399 6568.561
## 12 5250.789 5696.951
## 14 5183.349 5629.511
## 17 6122.399 6568.561
## 20 5250.789 5696.951
## 23 5183.349 5629.511
subset(af,series_id=="Y1" & horizon==2)
## series_id value timestamp method origin_timestamp forecast horizon
## 9 Y1 6158.68 1990 A 1988 5875.96 2
## 11 Y1 6158.68 1990 B 1988 6010.43 2
## 13 Y1 6158.68 1990 C 1988 5875.96 2
## 16 Y1 6876.58 1991 A 1989 6145.48 2
## 19 Y1 6876.58 1991 B 1989 6113.87 2
## 22 Y1 6876.58 1991 C 1989 6106.43 2
## lo90 hi90
## 9 5652.879 6099.041
## 11 5787.349 6233.511
## 13 5652.879 6099.041
## 16 6122.399 6568.561
## 19 5250.789 5696.951
## 22 5183.349 5629.511