healthyverse_tsa
Time Series Analysis, Modeling and Forecasting of the Healthyverse Packages ================ Steven P. Sanderson II, MPH - Date: 09 September, 2025
This analysis follows a Nested Modeltime Workflow.
Get Data
glimpse(downloads_tbl)
## Rows: 150,918
## Columns: 11
## $ date <date> 2020-11-23, 2020-11-23, 2020-11-23, 2020-11-23, 2020-11-23,…
## $ time <Period> 15H 36M 55S, 11H 26M 39S, 23H 34M 44S, 18H 39M 32S, 9H 0M…
## $ date_time <dttm> 2020-11-23 15:36:55, 2020-11-23 11:26:39, 2020-11-23 23:34:…
## $ size <int> 4858294, 4858294, 4858301, 4858295, 361, 4863722, 4864794, 4…
## $ r_version <chr> NA, "4.0.3", "3.5.3", "3.5.2", NA, NA, NA, NA, NA, NA, NA, N…
## $ r_arch <chr> NA, "x86_64", "x86_64", "x86_64", NA, NA, NA, NA, NA, NA, NA…
## $ r_os <chr> NA, "mingw32", "mingw32", "linux-gnu", NA, NA, NA, NA, NA, N…
## $ package <chr> "healthyR.data", "healthyR.data", "healthyR.data", "healthyR…
## $ version <chr> "1.0.0", "1.0.0", "1.0.0", "1.0.0", "1.0.0", "1.0.0", "1.0.0…
## $ country <chr> "US", "US", "US", "GB", "US", "US", "DE", "HK", "JP", "US", …
## $ ip_id <int> 2069, 2804, 78827, 27595, 90474, 90474, 42435, 74, 7655, 638…
The last day in the data set is 2025-09-07 23:35:42, the file was birthed on: 2024-08-07 07:35:44.428716, and at report knit time is -9516 hours old. Happy analyzing!
Now that we have our data lets take a look at it using the skimr
package.
skim(downloads_tbl)
| Name | downloads_tbl |
| Number of rows | 150918 |
| Number of columns | 11 |
| _______________________ | |
| Column type frequency: | |
| character | 6 |
| Date | 1 |
| numeric | 2 |
| POSIXct | 1 |
| Timespan | 1 |
| ________________________ | |
| Group variables | None |
Data summary
Variable type: character
| skim_variable | n_missing | complete_rate | min | max | empty | n_unique | whitespace |
|---|---|---|---|---|---|---|---|
| r_version | 109603 | 0.27 | 5 | 5 | 0 | 48 | 0 |
| r_arch | 109603 | 0.27 | 3 | 7 | 0 | 5 | 0 |
| r_os | 109603 | 0.27 | 7 | 15 | 0 | 23 | 0 |
| package | 0 | 1.00 | 7 | 13 | 0 | 8 | 0 |
| version | 0 | 1.00 | 5 | 17 | 0 | 62 | 0 |
| country | 12879 | 0.91 | 2 | 2 | 0 | 165 | 0 |
Variable type: Date
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| date | 0 | 1 | 2020-11-23 | 2025-09-07 | 2023-08-24 | 1743 |
Variable type: numeric
| skim_variable | n_missing | complete_rate | mean | sd | p0 | p25 | p50 | p75 | p100 | hist |
|---|---|---|---|---|---|---|---|---|---|---|
| size | 0 | 1 | 1130156.47 | 1504236.42 | 355 | 14701 | 306707 | 2365870 | 5677952 | ▇▁▂▁▁ |
| ip_id | 0 | 1 | 11251.97 | 21570.74 | 1 | 281 | 3058 | 12119 | 299146 | ▇▁▁▁▁ |
Variable type: POSIXct
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| date_time | 0 | 1 | 2020-11-23 09:00:41 | 2025-09-07 23:35:42 | 2023-08-24 18:03:48 | 93767 |
Variable type: Timespan
| skim_variable | n_missing | complete_rate | min | max | median | n_unique |
|---|---|---|---|---|---|---|
| time | 0 | 1 | 0 | 59 | 29 | 60 |
We can see that the following columns are missing a lot of data and for
us are most likely not useful anyways, so we will drop them
c(r_version, r_arch, r_os)
Plots
Now lets take a look at a time-series plot of the total daily downloads by package. We will use a log scale and place a vertical line at each version release for each package.
Now lets take a look at some time series decomposition graphs.
Feature Engineering
Now that we have our basic data and a shot of what it looks like, let’s
add some features to our data which can be very helpful in modeling.
Lets start by making a tibble that is aggregated by the day and
package, as we are going to be interested in forecasting the next 4
weeks or 28 days for each package. First lets get our base data.
##
## Call:
## stats::lm(formula = .formula, data = df)
##
## Residuals:
## Min 1Q Median 3Q Max
## -147.24 -36.07 -11.22 26.79 817.95
##
## Coefficients:
## Estimate Std. Error
## (Intercept) -1.781e+02 6.295e+01
## date 1.095e-02 3.333e-03
## lag(value, 1) 1.063e-01 2.374e-02
## lag(value, 7) 9.067e-02 2.453e-02
## lag(value, 14) 8.109e-02 2.457e-02
## lag(value, 21) 6.445e-02 2.465e-02
## lag(value, 28) 7.014e-02 2.456e-02
## lag(value, 35) 6.894e-02 2.466e-02
## lag(value, 42) 5.881e-02 2.479e-02
## lag(value, 49) 6.490e-02 2.466e-02
## month(date, label = TRUE).L -9.681e+00 5.102e+00
## month(date, label = TRUE).Q 3.173e+00 4.988e+00
## month(date, label = TRUE).C -1.346e+01 5.076e+00
## month(date, label = TRUE)^4 -6.838e+00 5.069e+00
## month(date, label = TRUE)^5 -1.098e+01 5.028e+00
## month(date, label = TRUE)^6 -4.245e+00 5.110e+00
## month(date, label = TRUE)^7 -7.623e+00 4.990e+00
## month(date, label = TRUE)^8 -3.512e+00 4.979e+00
## month(date, label = TRUE)^9 5.595e+00 4.968e+00
## month(date, label = TRUE)^10 3.767e+00 4.901e+00
## month(date, label = TRUE)^11 -2.910e+00 4.814e+00
## fourier_vec(date, type = "sin", K = 1, period = 7) -1.163e+01 2.269e+00
## fourier_vec(date, type = "cos", K = 1, period = 7) 7.669e+00 2.380e+00
## t value Pr(>|t|)
## (Intercept) -2.829 0.004723 **
## date 3.286 0.001036 **
## lag(value, 1) 4.479 8.01e-06 ***
## lag(value, 7) 3.696 0.000226 ***
## lag(value, 14) 3.300 0.000987 ***
## lag(value, 21) 2.614 0.009024 **
## lag(value, 28) 2.856 0.004339 **
## lag(value, 35) 2.796 0.005234 **
## lag(value, 42) 2.373 0.017775 *
## lag(value, 49) 2.632 0.008571 **
## month(date, label = TRUE).L -1.897 0.057939 .
## month(date, label = TRUE).Q 0.636 0.524794
## month(date, label = TRUE).C -2.651 0.008103 **
## month(date, label = TRUE)^4 -1.349 0.177550
## month(date, label = TRUE)^5 -2.183 0.029140 *
## month(date, label = TRUE)^6 -0.831 0.406260
## month(date, label = TRUE)^7 -1.528 0.126811
## month(date, label = TRUE)^8 -0.705 0.480755
## month(date, label = TRUE)^9 1.126 0.260159
## month(date, label = TRUE)^10 0.769 0.442262
## month(date, label = TRUE)^11 -0.605 0.545593
## fourier_vec(date, type = "sin", K = 1, period = 7) -5.125 3.31e-07 ***
## fourier_vec(date, type = "cos", K = 1, period = 7) 3.222 0.001295 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 58.77 on 1671 degrees of freedom
## (49 observations deleted due to missingness)
## Multiple R-squared: 0.2322, Adjusted R-squared: 0.2221
## F-statistic: 22.97 on 22 and 1671 DF, p-value: < 2.2e-16
NNS Forecasting
This is something I have been wanting to try for a while. The NNS
package is a great package for forecasting time series data.
library(NNS)
data_list <- base_data |>
select(package, value) |>
group_split(package)
data_list |>
imap(
\(x, idx) {
obj <- x
x <- obj |> pull(value) |> tail(7*52)
train_set_size <- length(x) - 56
pkg <- obj |> pluck(1) |> unique()
# sf <- NNS.seas(x, modulo = 7, plot = FALSE)$periods
seas <- t(
sapply(
1:25,
function(i) c(
i,
sqrt(
mean((
NNS.ARMA(x,
h = 28,
training.set = train_set_size,
method = "lin",
seasonal.factor = i,
plot=FALSE
) - tail(x, 28)) ^ 2)))
)
)
colnames(seas) <- c("Period", "RMSE")
sf <- seas[which.min(seas[, 2]), 1]
cat(paste0("Package: ", pkg, "\n"))
NNS.ARMA.optim(
variable = x,
h = 28,
training.set = train_set_size,
#seasonal.factor = seq(12, 60, 7),
seasonal.factor = sf,
pred.int = 0.95,
plot = TRUE
)
title(
sub = paste0("\n",
"Package: ", pkg, " - NNS Optimization")
)
}
)
## Package: healthyR
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 24 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 8.9552767925964"
## [1] "BEST method = 'lin' PATH MEMBER = c( 24 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 8.9552767925964"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 24 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 12.0888144741933"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 24 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 12.0888144741933"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 24 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 9.05915369504169"
## [1] "BEST method = 'both' PATH MEMBER = c( 24 )"
## [1] "BEST both OBJECTIVE FUNCTION = 9.05915369504169"
## Package: healthyR.ai
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 11.6839394132731"
## [1] "BEST method = 'lin' PATH MEMBER = c( 13 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 11.6839394132731"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 6.87157795156845"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 13 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 6.87157795156845"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 12.5753799943969"
## [1] "BEST method = 'both' PATH MEMBER = c( 13 )"
## [1] "BEST both OBJECTIVE FUNCTION = 12.5753799943969"
## Package: healthyR.data
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 6.08229264907597"
## [1] "BEST method = 'lin' PATH MEMBER = c( 13 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 6.08229264907597"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 7.79667313816308"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 13 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 7.79667313816308"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 13 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 7.71349406451732"
## [1] "BEST method = 'both' PATH MEMBER = c( 13 )"
## [1] "BEST both OBJECTIVE FUNCTION = 7.71349406451732"
## Package: healthyR.ts
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 22 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 7.65358404405505"
## [1] "BEST method = 'lin' PATH MEMBER = c( 22 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 7.65358404405505"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 22 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 3.44156780782819"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 22 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 3.44156780782819"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 22 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 3.17729648809361"
## [1] "BEST method = 'both' PATH MEMBER = c( 22 )"
## [1] "BEST both OBJECTIVE FUNCTION = 3.17729648809361"
## Package: healthyverse
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 7 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 14.9471657498516"
## [1] "BEST method = 'lin' PATH MEMBER = c( 7 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 14.9471657498516"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 7 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 7.61005519025815"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 7 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 7.61005519025815"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 7 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 11.3357451452868"
## [1] "BEST method = 'both' PATH MEMBER = c( 7 )"
## [1] "BEST both OBJECTIVE FUNCTION = 11.3357451452868"
## Package: RandomWalker
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 1 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 126.544841296215"
## [1] "BEST method = 'lin' PATH MEMBER = c( 1 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 126.544841296215"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 1 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 120.726723833397"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 1 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 120.726723833397"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 1 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 360.534882744396"
## [1] "BEST method = 'both' PATH MEMBER = c( 1 )"
## [1] "BEST both OBJECTIVE FUNCTION = 360.534882744396"
## Package: tidyAML
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 2 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 129.481687532028"
## [1] "BEST method = 'lin' PATH MEMBER = c( 2 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 129.481687532028"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 2 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 14.595026824739"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 2 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 14.595026824739"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 2 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 15.9051488697252"
## [1] "BEST method = 'both' PATH MEMBER = c( 2 )"
## [1] "BEST both OBJECTIVE FUNCTION = 15.9051488697252"
## Package: TidyDensity
## [1] "CURRNET METHOD: lin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'lin' , seasonal.factor = c( 9 ) ...)"
## [1] "CURRENT lin OBJECTIVE FUNCTION = 17.2375978709815"
## [1] "BEST method = 'lin' PATH MEMBER = c( 9 )"
## [1] "BEST lin OBJECTIVE FUNCTION = 17.2375978709815"
## [1] "CURRNET METHOD: nonlin"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'nonlin' , seasonal.factor = c( 9 ) ...)"
## [1] "CURRENT nonlin OBJECTIVE FUNCTION = 15.8104134005434"
## [1] "BEST method = 'nonlin' PATH MEMBER = c( 9 )"
## [1] "BEST nonlin OBJECTIVE FUNCTION = 15.8104134005434"
## [1] "CURRNET METHOD: both"
## [1] "COPY LATEST PARAMETERS DIRECTLY FOR NNS.ARMA() IF ERROR:"
## [1] "NNS.ARMA(... method = 'both' , seasonal.factor = c( 9 ) ...)"
## [1] "CURRENT both OBJECTIVE FUNCTION = 17.3408213989762"
## [1] "BEST method = 'both' PATH MEMBER = c( 9 )"
## [1] "BEST both OBJECTIVE FUNCTION = 17.3408213989762"
## [[1]]
## NULL
##
## [[2]]
## NULL
##
## [[3]]
## NULL
##
## [[4]]
## NULL
##
## [[5]]
## NULL
##
## [[6]]
## NULL
##
## [[7]]
## NULL
##
## [[8]]
## NULL
Pre-Processing
Now we are going to do some basic pre-processing.
data_padded_tbl <- base_data %>%
pad_by_time(
.date_var = date,
.pad_value = 0
)
# Get log interval and standardization parameters
log_params <- liv(data_padded_tbl$value, limit_lower = 0, offset = 1, silent = TRUE)
limit_lower <- log_params$limit_lower
limit_upper <- log_params$limit_upper
offset <- log_params$offset
data_liv_tbl <- data_padded_tbl %>%
# Get log interval transform
mutate(value_trans = liv(value, limit_lower = 0, offset = 1, silent = TRUE)$log_scaled)
# Get Standardization Params
std_params <- standard_vec(data_liv_tbl$value_trans, silent = TRUE)
std_mean <- std_params$mean
std_sd <- std_params$sd
data_transformed_tbl <- data_liv_tbl %>%
# get standardization
mutate(value_trans = standard_vec(value_trans, silent = TRUE)$standard_scaled) %>%
select(-value)
Since this is panel data we can follow one of two different modeling strategies. We can search for a global model in the panel data or we can use nested forecasting finding the best model for each of the time series. Since we only have 5 panels, we will use nested forecasting.
To do this we will use the nest_timeseries and
split_nested_timeseries functions to create a nested tibble.
horizon <- 4*7
nested_data_tbl <- data_transformed_tbl %>%
# 1. Extending: We'll predict n days into the future.
extend_timeseries(
.id_var = package,
.date_var = date,
.length_future = horizon
) %>%
# 2. Nesting: We'll group by id, and create a future dataset
# that forecasts n days of extended data and
# an actual dataset that contains n*2 days
nest_timeseries(
.id_var = package,
.length_future = horizon
#.length_actual = horizon*2
) %>%
# 3. Splitting: We'll take the actual data and create splits
# for accuracy and confidence interval estimation of n das (test)
# and the rest is training data
split_nested_timeseries(
.length_test = horizon
)
nested_data_tbl
## # A tibble: 9 × 4
## package .actual_data .future_data .splits
## <fct> <list> <list> <list>
## 1 healthyR.data <tibble [1,735 × 2]> <tibble [28 × 2]> <split [1707|28]>
## 2 healthyR <tibble [1,726 × 2]> <tibble [28 × 2]> <split [1698|28]>
## 3 healthyR.ts <tibble [1,672 × 2]> <tibble [28 × 2]> <split [1644|28]>
## 4 healthyverse <tibble [1,643 × 2]> <tibble [28 × 2]> <split [1615|28]>
## 5 healthyR.ai <tibble [1,468 × 2]> <tibble [28 × 2]> <split [1440|28]>
## 6 TidyDensity <tibble [1,319 × 2]> <tibble [28 × 2]> <split [1291|28]>
## 7 tidyAML <tibble [926 × 2]> <tibble [28 × 2]> <split [898|28]>
## 8 RandomWalker <tibble [349 × 2]> <tibble [28 × 2]> <split [321|28]>
## 9 <NA> <tibble [7 × 2]> <tibble [28 × 2]> <split [0|7]>
Now it is time to make some recipes and models using the modeltime workflow.
Modeltime Workflow
Recipe Object
recipe_base <- recipe(
value_trans ~ date
, data = extract_nested_test_split(nested_data_tbl)
)
recipe_base
recipe_date <- recipe_base %>%
step_mutate(date = as.numeric(date))
Models
# Models ------------------------------------------------------------------
# Auto ARIMA --------------------------------------------------------------
model_spec_arima_no_boost <- arima_reg() %>%
set_engine(engine = "auto_arima")
wflw_auto_arima <- workflow() %>%
add_recipe(recipe = recipe_base) %>%
add_model(model_spec_arima_no_boost)
# NNETAR ------------------------------------------------------------------
model_spec_nnetar <- nnetar_reg(
mode = "regression"
, seasonal_period = "auto"
) %>%
set_engine("nnetar")
wflw_nnetar <- workflow() %>%
add_recipe(recipe = recipe_base) %>%
add_model(model_spec_nnetar)
# TSLM --------------------------------------------------------------------
model_spec_lm <- linear_reg() %>%
set_engine("lm")
wflw_lm <- workflow() %>%
add_recipe(recipe = recipe_base) %>%
add_model(model_spec_lm)
# MARS --------------------------------------------------------------------
model_spec_mars <- mars(mode = "regression") %>%
set_engine("earth")
wflw_mars <- workflow() %>%
add_recipe(recipe = recipe_base) %>%
add_model(model_spec_mars)
Nested Modeltime Tables
nested_modeltime_tbl <- modeltime_nested_fit(
# Nested Data
nested_data = nested_data_tbl,
control = control_nested_fit(
verbose = TRUE,
allow_par = FALSE
),
# Add workflows
wflw_auto_arima,
wflw_lm,
wflw_mars,
wflw_nnetar
)
nested_modeltime_tbl <- nested_modeltime_tbl[!is.na(nested_modeltime_tbl$package),]
Model Accuracy
nested_modeltime_tbl %>%
extract_nested_test_accuracy() %>%
filter(!is.na(package)) %>%
knitr::kable()
| package | .model_id | .model_desc | .type | mae | mape | mase | smape | rmse | rsq |
|---|---|---|---|---|---|---|---|---|---|
| healthyR.data | 1 | ARIMA | Test | 0.8426767 | 127.52481 | 1.0287249 | 160.28953 | 1.0065555 | 0.0079960 |
| healthyR.data | 2 | LM | Test | 0.8941997 | 191.62638 | 1.0916233 | 152.51250 | 1.0182284 | 0.1611050 |
| healthyR.data | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| healthyR.data | 4 | NNAR | Test | 0.8107353 | 100.61609 | 0.9897314 | 183.10194 | 0.9982203 | 0.0019347 |
| healthyR | 1 | ARIMA | Test | 0.6659630 | 105.46053 | 0.8348659 | 169.83244 | 0.8191942 | 0.0029507 |
| healthyR | 2 | LM | Test | 0.6821286 | 112.11783 | 0.8551315 | 174.68849 | 0.8114274 | 0.1035413 |
| healthyR | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| healthyR | 4 | NNAR | Test | 0.6582019 | 111.70376 | 0.8251364 | 157.34955 | 0.7989028 | 0.0537168 |
| healthyR.ts | 1 | ARIMA | Test | 0.8302963 | 142.00948 | 0.9453313 | 123.48903 | 1.0494157 | 0.0077396 |
| healthyR.ts | 2 | LM | Test | 0.8655259 | 133.16167 | 0.9854417 | 132.76118 | 1.0545217 | 0.1772334 |
| healthyR.ts | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| healthyR.ts | 4 | NNAR | Test | 0.8612081 | 87.42003 | 0.9805258 | 157.88271 | 1.0801306 | 0.0304579 |
| healthyverse | 1 | ARIMA | Test | 0.7619207 | 131.47750 | 1.1727421 | 103.05855 | 0.9002618 | 0.2417428 |
| healthyverse | 2 | LM | Test | 0.7939527 | 167.26367 | 1.2220455 | 99.30881 | 0.9227284 | 0.4003052 |
| healthyverse | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| healthyverse | 4 | NNAR | Test | 0.7938664 | 110.22277 | 1.2219127 | 116.25526 | 0.9578655 | 0.2033114 |
| healthyR.ai | 1 | ARIMA | Test | 0.6850467 | 84.54703 | 1.0876992 | 143.14192 | 0.8265603 | 0.1947200 |
| healthyR.ai | 2 | LM | Test | 0.7354183 | 93.15843 | 1.1676779 | 144.45571 | 0.8717839 | 0.2546153 |
| healthyR.ai | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| healthyR.ai | 4 | NNAR | Test | 0.6888056 | 86.24269 | 1.0936675 | 145.97591 | 0.8434572 | 0.0623689 |
| TidyDensity | 1 | ARIMA | Test | 0.4857572 | 235.95346 | 0.8125713 | 87.22950 | 0.6214379 | 0.1330548 |
| TidyDensity | 2 | LM | Test | 0.5364293 | 245.92336 | 0.8973352 | 92.20318 | 0.6556847 | 0.0208821 |
| TidyDensity | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| TidyDensity | 4 | NNAR | Test | 0.5053273 | 137.72727 | 0.8453079 | 109.74415 | 0.6805769 | 0.0793760 |
| tidyAML | 1 | ARIMA | Test | 0.6450618 | 213.16429 | 1.2752408 | 114.12181 | 0.7748191 | 0.0156459 |
| tidyAML | 2 | LM | Test | 0.6525595 | 274.69780 | 1.2900633 | 107.57965 | 0.7685128 | 0.4150717 |
| tidyAML | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| tidyAML | 4 | NNAR | Test | 0.6367590 | 218.75326 | 1.2588269 | 111.41655 | 0.7645486 | 0.0007219 |
| RandomWalker | 1 | ARIMA | Test | 1.0980332 | 122.35208 | 0.6803580 | 174.30804 | 1.3070925 | 0.0004890 |
| RandomWalker | 2 | LM | Test | 1.0706294 | 120.85628 | 0.6633782 | 175.81788 | 1.2663354 | 0.0058968 |
| RandomWalker | 3 | NULL | NA | NA | NA | NA | NA | NA | NA |
| RandomWalker | 4 | NNAR | Test | 1.2366872 | 190.82965 | 0.7662701 | 156.43693 | 1.3915555 | 0.0000854 |
Plot Models
nested_modeltime_tbl %>%
extract_nested_test_forecast() %>%
group_by(package) %>%
plot_modeltime_forecast(
.interactive = FALSE,
.conf_interval_show = FALSE,
.facet_scales = "free"
) +
theme_minimal() +
theme(legend.position = "bottom")
Best Model
best_nested_modeltime_tbl <- nested_modeltime_tbl %>%
modeltime_nested_select_best(
metric = "rmse",
minimize = TRUE,
filter_test_forecasts = TRUE
)
best_nested_modeltime_tbl %>%
extract_nested_best_model_report()
## # Nested Modeltime Table
##
## # A tibble: 8 × 10
## package .model_id .model_desc .type mae mape mase smape rmse rsq
## <fct> <int> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 healthyR.da… 4 NNAR Test 0.811 101. 0.990 183. 0.998 1.93e-3
## 2 healthyR 4 NNAR Test 0.658 112. 0.825 157. 0.799 5.37e-2
## 3 healthyR.ts 1 ARIMA Test 0.830 142. 0.945 123. 1.05 7.74e-3
## 4 healthyverse 1 ARIMA Test 0.762 131. 1.17 103. 0.900 2.42e-1
## 5 healthyR.ai 1 ARIMA Test 0.685 84.5 1.09 143. 0.827 1.95e-1
## 6 TidyDensity 1 ARIMA Test 0.486 236. 0.813 87.2 0.621 1.33e-1
## 7 tidyAML 4 NNAR Test 0.637 219. 1.26 111. 0.765 7.22e-4
## 8 RandomWalker 2 LM Test 1.07 121. 0.663 176. 1.27 5.90e-3
best_nested_modeltime_tbl %>%
extract_nested_test_forecast() %>%
#filter(!is.na(.model_id)) %>%
group_by(package) %>%
plot_modeltime_forecast(
.interactive = FALSE,
.conf_interval_alpha = 0.2,
.facet_scales = "free"
) +
theme_minimal() +
theme(legend.position = "bottom")
Refitting and Future Forecast
Now that we have the best models, we can make our future forecasts.
nested_modeltime_refit_tbl <- best_nested_modeltime_tbl %>%
modeltime_nested_refit(
control = control_nested_refit(verbose = TRUE)
)
nested_modeltime_refit_tbl
## # Nested Modeltime Table
##
## # A tibble: 8 × 5
## package .actual_data .future_data .splits .modeltime_tables
## <fct> <list> <list> <list> <list>
## 1 healthyR.data <tibble> <tibble> <split [1707|28]> <mdl_tm_t [1 × 5]>
## 2 healthyR <tibble> <tibble> <split [1698|28]> <mdl_tm_t [1 × 5]>
## 3 healthyR.ts <tibble> <tibble> <split [1644|28]> <mdl_tm_t [1 × 5]>
## 4 healthyverse <tibble> <tibble> <split [1615|28]> <mdl_tm_t [1 × 5]>
## 5 healthyR.ai <tibble> <tibble> <split [1440|28]> <mdl_tm_t [1 × 5]>
## 6 TidyDensity <tibble> <tibble> <split [1291|28]> <mdl_tm_t [1 × 5]>
## 7 tidyAML <tibble> <tibble> <split [898|28]> <mdl_tm_t [1 × 5]>
## 8 RandomWalker <tibble> <tibble> <split [321|28]> <mdl_tm_t [1 × 5]>
nested_modeltime_refit_tbl %>%
extract_nested_future_forecast() %>%
mutate(across(.value:.conf_hi, .fns = ~ standard_inv_vec(
x = .,
mean = std_mean,
sd = std_sd
)$standard_inverse_value)) %>%
mutate(across(.value:.conf_hi, .fns = ~ liiv(
x = .,
limit_lower = limit_lower,
limit_upper = limit_upper,
offset = offset
)$rescaled_v)) %>%
group_by(package) %>%
plot_modeltime_forecast(
.interactive = FALSE,
.conf_interval_alpha = 0.2,
.facet_scales = "free"
) +
theme_minimal() +
theme(legend.position = "bottom")
