What does this package do ?

The autoTS package provides a high level interface for univariate time series predictions. It implements many algorithms, most of them provided by the forecast package. The main goals of the package are :

• Simplify the preparation of the time series ;
• Train the algorithms and compare their results, to chose the best one ;
• Gather the results in a final tidy dataframe

What are the inputs ?

The package is designed to work on one time series at a time. Parallel calculations can be put on top of it (see example below). The user has to provide 2 simple vectors :

• One with the dates (s.t. the lubridate package can parse them)
• The second with the corresponding values

Warnings

This package implements each algorithm with a unique parametrization, meaning that the user cannot tweak the algorithms (eg modify SARIMA specfic parameters).

Data preparation

First, we have to clean the data (not too ugly though). First thing is to convert the TIME column into a well known date format that lubridate can handle. In this example, the yq function can parse the date without modification of the column. Then, we have to remove the blank in the values that separates thousands… Finally, we only keep data since 2000 and the unadjusted series in current prices.

After that, we should get one time series per country

dat <- mutate(dat,dates=yq(as.character(TIME)),
              values = as.numeric(stringr::str_remove(Value," "))) %>% 
  filter(year(dates)>=2000 & 
           S_ADJ=="Unadjusted data (i.e. neither seasonally adjusted nor calendar adjusted data)" &
           UNIT == "Current prices, million euro")

filter(dat,GEO %in% c("France","Austria")) %>% 
  ggplot(aes(dates,values,color=GEO)) + geom_line() + theme_minimal() +
  labs(title="GDP of (completely) random countries")

Now we’re good to go !

Prediction on a random country

Let’s see how to use the package on one time series :

• Extract dates and values of the time series you want to work on
• Create the object containing all you need afterwards
• Train algo and determine which one is the best (over the last known year)
• Implement the best algorithm on full data

ex1 <- filter(dat,GEO=="France") 
preparedTS <- prepare.ts(ex1$dates,ex1$values,"quarter")

## What is in this new object ?
str(preparedTS)

## List of 4
##  $ obj.ts    : Time-Series [1:77] from 2000 to 2019: 363007 369185 362905 383489 380714 ...
##  $ obj.df    :'data.frame':  77 obs. of  2 variables:
##   ..$ dates: Date[1:77], format: "2000-01-01" "2000-04-01" ...
##   ..$ val  : num [1:77] 363007 369185 362905 383489 380714 ...
##  $ freq.num  : num 4
##  $ freq.alpha: chr "quarter"

plot.ts(preparedTS$obj.ts)

ggplot(preparedTS$obj.df,aes(dates,val)) + geom_line() + theme_minimal()

Get the best algorithm for this time series :

## What is the best model for prediction ?
best.algo <- getBestModel(ex1$dates,ex1$values,"quarter",graph = F)
names(best.algo)

## [1] "prepedTS"     "best"         "train.errors" "res.train"    "algos"       
## [6] "graph.train"

print(paste("The best algorithm is",best.algo$best))

## [1] "The best algorithm is my.ets"

best.algo$graph.train

You find in the result of this function :

• The name of the best model
• The errors of each algorithm on the test set
• The graphic of the train step
• The prepared time series
• The list of used algorithm (that you can customize)

The result of this function can be used as direct input of the my.prediction function

## Build the predictions
final.pred <- my.predictions(bestmod = best.algo)
tail(final.pred,24)

## # A tibble: 24 × 4
##    dates      type  actual.value   ets
##    <date>     <chr>        <dbl> <dbl>
##  1 2016-04-01 <NA>        560873    NA
##  2 2016-07-01 <NA>        546383    NA
##  3 2016-10-01 <NA>        572752    NA
##  4 2017-01-01 <NA>        565221    NA
##  5 2017-04-01 <NA>        573720    NA
##  6 2017-07-01 <NA>        563671    NA
##  7 2017-10-01 <NA>        592453    NA
##  8 2018-01-01 <NA>        580884    NA
##  9 2018-04-01 <NA>        586869    NA
## 10 2018-07-01 <NA>        577904    NA
## # ℹ 14 more rows

ggplot(final.pred) + geom_line(aes(dates,actual.value),color="black") + 
  geom_line(aes_string("dates",stringr::str_remove(best.algo$best,"my."),linetype="type"),color="red") +
  theme_minimal() 

Not too bad, right ?

Prepare data

suppressPackageStartupMessages(library(tidyr))
dat.wide <- select(dat,GEO,dates,values) %>% 
  group_by(dates) %>% 
  spread(key = "GEO",value = "values")
head(dat.wide)

## # A tibble: 6 × 45
## # Groups:   dates [6]
##   dates      Albania Austria Belgium `Bosnia and Herzegovina` Bulgaria Croatia
##   <date>       <dbl>   <dbl>   <dbl>                    <dbl>    <dbl>   <dbl>
## 1 2000-01-01      NA  50422.   62261                       NA    2941.   5266.
## 2 2000-04-01      NA  53180.   65046                       NA    3252.   5811 
## 3 2000-07-01      NA  53881.   62754                       NA    4015.   6409.
## 4 2000-10-01      NA  56123.   68161                       NA    4103.   6113 
## 5 2001-01-01      NA  52911.   64318                       NA    3284.   5777.
## 6 2001-04-01      NA  54994.   67537                       NA    3669.   6616.
## # ℹ 38 more variables: Cyprus <dbl>, Czechia <dbl>, Denmark <dbl>,
## #   Estonia <dbl>, `Euro area (12 countries)` <dbl>,
## #   `Euro area (19 countries)` <dbl>,
## #   `Euro area (EA11-2000, EA12-2006, EA13-2007, EA15-2008, EA16-2010, EA17-2013, EA18-2014, EA19)` <dbl>,
## #   `European Union - 15 countries (1995-2004)` <dbl>,
## #   `European Union - 27 countries (from 2019)` <dbl>,
## #   `European Union - 28 countries` <dbl>, Finland <dbl>, France <dbl>, …

Compute bulk predictions

Note : The following code is not executed for this vignette but does work (you can try it at home)

library(doParallel)
pipeline <- function(dates,values)
{
  pred <- getBestModel(dates,values,"quarter",graph = F)  %>%
    my.predictions()
  return(pred)
}
doMC::registerDoMC(parallel::detectCores()-1) # parallel backend (for UNIX)

system.time({
  res <- foreach(ii=2:ncol(dat.wide),.packages = c("dplyr","autoTS")) %dopar%
  pipeline(dat.wide$dates,pull(dat.wide,ii))
})
names(res) <- colnames(dat.wide)[-1]
str(res)

There is no free lunch…

There is no best algorithm in general ⇒ depends on the data ! Likewise, this is not executed in this vignette, but works if you want to replicate it.

sapply(res,function(xx) colnames(select(xx,-dates,-type,-actual.value)) ) %>% table()
sapply(res,function(xx) colnames(select(xx,-dates,-type,-actual.value)) )