convert numeric date format into correct date format using auto.arima() by group in R

Question

Let's work with this data sample

timeseries<-structure(list(Data = structure(c(10L, 14L, 18L, 22L, 26L, 29L, 
32L, 35L, 38L, 1L, 4L, 7L, 11L, 15L, 19L, 23L, 27L, 30L, 33L, 
36L, 39L, 2L, 5L, 8L, 12L, 16L, 20L, 24L, 28L, 31L, 34L, 37L, 
40L, 3L, 6L, 9L, 13L, 17L, 21L, 25L), .Label = c("01.01.2018", 
"01.01.2019", "01.01.2020", "01.02.2018", "01.02.2019", "01.02.2020", 
"01.03.2018", "01.03.2019", "01.03.2020", "01.04.2017", "01.04.2018", 
"01.04.2019", "01.04.2020", "01.05.2017", "01.05.2018", "01.05.2019", 
"01.05.2020", "01.06.2017", "01.06.2018", "01.06.2019", "01.06.2020", 
"01.07.2017", "01.07.2018", "01.07.2019", "01.07.2020", "01.08.2017", 
"01.08.2018", "01.08.2019", "01.09.2017", "01.09.2018", "01.09.2019", 
"01.10.2017", "01.10.2018", "01.10.2019", "01.11.2017", "01.11.2018", 
"01.11.2019", "01.12.2017", "01.12.2018", "01.12.2019"), class = "factor"), 
    client = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L, 1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 
    2L, 2L, 2L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 
    1L), .Label = c("Horns", "Kornev"), class = "factor"), stuff = structure(c(1L, 
    1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L, 3L, 3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 2L, 2L, 2L, 2L, 
    2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L), .Label = c("chickens", 
    "hooves", "Oysters"), class = "factor"), Sales = c(374L, 
    12L, 120L, 242L, 227L, 268L, 280L, 419L, 12L, 172L, 336L, 
    117L, 108L, 150L, 90L, 117L, 116L, 146L, 120L, 211L, 213L, 
    67L, 146L, 118L, 152L, 122L, 201L, 497L, 522L, 65L, 268L, 
    441L, 247L, 348L, 445L, 477L, 62L, 226L, 476L, 306L)), .Names = c("Data", 
"client", "stuff", "Sales"), class = "data.frame", row.names = c(NA, 
-40L))

I want to perform forecast using auto.arima by group

# first the grouping variable
timeseries$group <- paste0(timeseries$client,timeseries$stuff)

# now the list
listed <- split(timeseries,timeseries$group)

library("forecast")
library("lubridate")

listed_ts <- lapply(listed,
                    function(x) ts(x[["Sales"]], start = ymd("2017-01-04"), frequency = 12)  ) 

listed_ts

listed_arima <- lapply(listed_ts,function(x) auto.arima(x) )
#Now the forecast for each arima:
listed_forecast <- lapply(listed_arima,function(x) forecast(x,2) )
listed_forecast
do.call(rbind,listed_forecast)

and after listed_forecast i get the next output

  Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov
    17170 374  12 120 242 227 268 280 419  12 172 336

    $Hornshooves
          Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
    17170 497 522  65 268 441 247 348 445 477  62 226 476
    17171 306                                            

    $KornevOysters
          Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
    17170 117 108 150  90 117 116 146 120 211 213  67 146
    17171 118 152 122 201

17170 is not correct date format

So as.numeric(ymd("2017-01-04")) shows us 17170

How to get output with correct date format? i want like this with normal date format

$Hornschickens
          Point Forecast    Lo 80    Hi 80     Lo 95    Hi 95
Dec 2017       223.8182 50.98365 396.6527 -40.50942 488.1458
Jan 2018       223.8182 50.98365 396.6527 -40.50942 488.1458

So i suppose that 17170 in forecast mean 2017,17171=2018 But any trouble to convert it in reading format

i see next logic Is it possible to do automatic conversion of these numeric values to dates

where

1970-01-01 (Y-m-d) is 0

1970-01-02 is 1

"2018-12-25" is 17890

but when aggregated by month input data like in my example

dec 17890 so simple convert to dec 2018 cause 17890 in the range of December 2018.

Julius Vainora · Accepted Answer · 2018-12-25T14:45:51.573

1

The start parameter in listed_ts isn't supposed to be a date:

start --- the time of the first observation. Either a single number or a vector of two integers, which specify a natural time unit and a (1-based) number of samples into the time unit. See the examples for the use of the second form.

(From ?ts.) So, instead using start = c(2017, 1) does the job.

edited Dec 25 '18 at 14:45

answered Dec 25 '18 at 14:28

Julius Vainora

47,421
9
90
102

may i ask you help me in related topic? https://stackoverflow.com/questions/53924851/display-predicted-values-for-initial-data-using-auto-arima-in-r – psysky Dec 25 '18 at 18:51

convert numeric date format into correct date format using auto.arima() by group in R

1 Answers1