tests/testthat/test-algo-temporal_hierarchy.R

In modeltime: The Tidymodels Extension for Time Series Modeling

# ---- HIERARCHICAL REG ----
context("TEST temporal_hierarchy: thief")


# TESTS
test_that("thief", {

    skip_on_cran()

    # temporal_hierarchy: thief, Test Model Fit Object ----

    # SETUP ----

    # Data
    m750 <- m4_monthly %>% filter(id == "M750")

    # Split Data 80/20
    splits <- initial_time_split(m750, prop = 0.9)

    # Model Spec
    model_spec <- temporal_hierarchy() %>%
        set_engine("thief")


    # HIERARCHICAL PARSNIP ----

    # * NO XREGS ----

    # Fit Spec
    model_fit <- model_spec %>%
        fit(log(value) ~ date, data = training(splits))

    # Predictions
    predictions_tbl <- model_fit %>%
        modeltime_calibrate(testing(splits)) %>%
        modeltime_forecast(new_data = testing(splits))

    testthat::expect_s3_class(model_fit$fit, "temporal_hier_fit_impl")

    # $fit

    testthat::expect_s3_class(model_fit$fit$models$model_1, "forecast")

    testthat::expect_s3_class(model_fit$fit$data, "tbl_df")

    testthat::expect_equal(names(model_fit$fit$data)[1], "date")

    testthat::expect_true(is.null(model_fit$fit$extras$xreg_recipe))

    # $preproc

    testthat::expect_equal(model_fit$preproc$y_var, "value")


    # Structure
    testthat::expect_identical(nrow(testing(splits)), nrow(predictions_tbl))
    testthat::expect_identical(testing(splits)$date, predictions_tbl$.index)

    # Out-of-Sample Accuracy Tests

    resid <- testing(splits)$value - exp(predictions_tbl$.value)

    # - Max Error less than 1500
    testthat::expect_lte(max(abs(resid)), 320)

    # - MAE less than 700
    testthat::expect_lte(mean(abs(resid)), 100)




    # ---- ETS WORKFLOWS ----

    # Model Spec
    model_spec <- temporal_hierarchy() %>%
        set_engine("thief")

    # Recipe spec
    recipe_spec <- recipe(value ~ date, data = training(splits)) %>%
        step_log(value, skip = FALSE)

    # Workflow
    wflw <- workflow() %>%
        add_recipe(recipe_spec) %>%
        add_model(model_spec)

    wflw_fit <- wflw %>%
        fit(training(splits))

    # Forecast
    predictions_tbl <- wflw_fit %>%
        modeltime_calibrate(testing(splits)) %>%
        modeltime_forecast(new_data = testing(splits),
                           actual_data = training(splits)) %>%
        mutate_at(vars(.value), exp)

    # Tests

    testthat::expect_s3_class(wflw_fit$fit$fit$fit, "temporal_hier_fit_impl")

    # $fit

    testthat::expect_s3_class(wflw_fit$fit$fit$fit$models$model_1, "forecast")

    testthat::expect_s3_class(wflw_fit$fit$fit$fit$data, "tbl_df")

    testthat::expect_equal(names(wflw_fit$fit$fit$fit$data)[1], "date")

    testthat::expect_true(is.null(wflw_fit$fit$fit$fit$extras$xreg_recipe))

    # $preproc
    mld <- wflw_fit %>% workflows::extract_mold()
    testthat::expect_equal(names(mld$outcomes), "value")


    full_data <- bind_rows(training(splits), testing(splits))

    # Structure
    testthat::expect_identical(nrow(full_data), nrow(predictions_tbl))
    testthat::expect_identical(full_data$date, predictions_tbl$.index)

    # Out-of-Sample Accuracy Tests
    predictions_tbl <- predictions_tbl %>% filter(.key == "prediction")
    resid <- testing(splits)$value - predictions_tbl$.value

    # - Max Error less than 1500
    testthat::expect_lte(max(abs(resid)), 320)

    # - MAE less than 700
    testthat::expect_lte(mean(abs(resid)), 100)

})