STOP 25 / 31Three approaches. Different trade-offs.
Layout
STOP 25 · More models

Three approaches. Different trade-offs.

EXPLORE~9 min
Back to route map
  • All three approaches can model complex seasonal sales, but they divide the problem differently.
  • Prophet combines trend, Fourier seasonality, holidays, and optional regressors.
  • DHR uses Fourier regression with ARIMA errors.
  • MSTL + ETS first removes estimated seasonal components, forecasts the adjusted series, then adds projected seasons back.
  • Use the table to compare the actual assumptions—not just the names.
Prophet vs DHR vs MSTL + ETS: concepts and Northstar examples. Illustrative Northstar examples, not measured business results.
ConceptWhat it meansNorthstar example
Representation is not the same as forecasting behavior.
  • Standard Prophet and standard DHR usually use fixed Fourier coefficients for a fitted seasonal shape.
  • MSTL can estimate seasonality that changes through the historical sample, but the common seasonal-naïve projection repeats its last estimated cycle.
  • None automatically knows how tomorrow’s shopping habits will change.
  • A historical seasonal plot that changes over time does not prove Northstar's pipeline predicts future changes in that pattern.
  • Check each method's actual projection rule.
Match strengths to the failure you need to fix.
  • Prophet provides a convenient trend/changepoint and holiday interface.
  • DHR makes known drivers and autocorrelated errors explicit.
  • MSTL + ETS provides an interpretable decomposition and a modular forecast for the adjusted series.
  • These are useful starting points, not performance rankings.

If Northstar's adjusted residuals still have memory, DHR's ARIMA errors are a motivated candidate; that is a reason to test it, not declare it the winner.

Compare the full pipelines fairly.
  • Use identical origins, horizons, training data, known-ahead information, and preprocessing rules.
  • Tune Fourier orders, changepoint flexibility, smoothing windows, and ETS/ARIMA choices within validation.
  • Evaluate ordinary days, event windows, bias, interval coverage, and operating cost against the same simple baselines.

Compare Prophet, DHR, and MSTL + ETS on the same Northstar forecast origins, horizons, calendars, and baselines, with every pipeline refit inside training windows.

TABLE 25.1

Prophet vs DHR vs MSTL + ETS

WHAT THIS SHOWS

Compare Prophet, DHR, and MSTL + ETS by mechanics, assumptions, and use case—not invented performance scores.

The question it answers: Which trade-offs matter for Northstar's actual decision?

Compare the standard versions of these pipelines. Extensions and implementation defaults can change their behavior.

Prophet vs DHR vs MSTL + ETS: assumptions, strengths, and limitations
CompareProphetDHRMSTL + ETS
What it doesFits a trend/changepoint model plus Fourier seasonality, holiday effects, and optional regressors.Fits regression with Fourier seasonal terms and optional known drivers; an ARIMA error process models remaining serial dependence.Decomposes multiple seasonal patterns; fits nonseasonal ETS to the adjusted series; projects seasonal components and recombines.
StrengthsConvenient interface for trend changes, calendars, and holiday windows. A practical candidate for business series with strong patterns and enough history.Explicit calendar/driver effects, smooth long or multiple seasonalities, and autocorrelated errors in one framework. Fourier orders control seasonal flexibility.Readable component breakdown; can estimate evolving historical seasonal shapes. Modular: choose the adjusted-series model and seasonal projection separately.
WeaknessesTrend extrapolation can be implausible; too much changepoint flexibility can overfit. Standard errors do not have an explicit ARIMA correlation model. Not automatically accurate.Seasonal periods, Fourier orders, regressors, and error structure need care. Too many harmonics overfit; fixed seasonal coefficients may miss evolving shapes. ARIMA fitting can add cost.Needs enough cycles and sensible smoothing settings. Boundary estimates can be fragile. Basic seasonal-naïve projection repeats the last seasonal shape; moving events need extra handling.
Trend & short-run memoryTrend is often piecewise linear, optionally with a saturating form. Standard Prophet does not explicitly model residual autocorrelation with AR/MA terms.Trend may be represented by regression terms and/or the ARIMA error specification. Lagged values and innovations can capture remaining short-run dependence.ETS forecasts level/trend behavior in the seasonally adjusted series. ETS innovations are not the same as choosing a general ARIMA error structure.
Changing seasonal patternsStandard seasonal coefficients are fixed within a fit. Refitting or deliberately adding interactions/conditional effects can adapt the model.Standard Fourier coefficients are fixed within a fit. Time-varying coefficients, interactions, or refitting are additional modeling choices.Finite smoothing windows allow changes across the historical sample. The common future rule repeats final estimated cycles; it does not extrapolate new seasonal changes by itself.
Black Friday & moving eventsSupply the actual holiday dates and lead/lag windows. The holiday interface is convenient, but the lift and future promotion assumptions still require evidence.Add event-relative indicators or windows alongside seasonal waves. Calendar dates are known ahead; future discounts or spend must be known, forecast, or scenarized.A fixed seasonal period is not an event calendar. Adjust/model the event separately using training data, forecast its effect, and add it back without double counting.
External predictorsAdditional regressors can be supplied. Their future values must be available or supplied as scenarios.A natural regression extension: price, promotions, weather, or availability indicators where justified. Unknown future inputs need forecasts or scenarios.Plain MSTL + ETS has no general external-regressor layer. Incorporating drivers requires an adjustment or a different/hybrid forecasting pipeline.
UncertaintyIntervals depend on model assumptions, fitting choices, and how future trend changes are simulated. Validate coverage; displayed bands are not guarantees.ARIMA errors provide conditional forecast uncertainty. Fixed future inputs and fitted coefficients may omit predictor or parameter uncertainty; validate coverage.Simple implementations often condition on estimated seasonal paths. Adjusted-series intervals alone can omit decomposition and future seasonal uncertainty; validate the whole pipeline.
What to tune and testChangepoint flexibility, seasonal orders/regularization, and holiday effects. Check trend plausibility and event errors.Fourier pairs for each period, driver specification, differencing and ARIMA orders. Check residual dependence and multi-step errors.Periods, smoothing/robustness settings, ETS structure, and seasonal extrapolation. Refit decomposition at every backtest origin.

Prophet and DHR both let you represent known calendar effects directly. MSTL + ETS remains a valid candidate if you deliberately add an event-adjustment layer. A sharp, moving sales event should not be left for smooth annual seasonality to discover automatically.

A fair test, not a popularity contest.

Same forecast origins, horizons, known-ahead inputs, preprocessing rules, and baselines. Check MAE/loss, bias, event-specific errors, interval coverage, runtime, and maintenance. There is no universal best model.

Fine print & assumptions

Decision guide: validate each candidate on the actual business decision and future-data tests.

CHECK YOURSELF · CHOOSE AN ANSWER

Northstar’s seasonality is mostly explained, but forecast errors remain autocorrelated. Which comparison is most directly motivated?

Your learning note

Private · just for you