: 01494 785574 (UK)

Prophecy™ Statistical forecasting engine

Dual forecasting engines!

Prophecy has had a built-in statistical forecasting engine from day 1, providing core time-series modelling without the need for statistical knowledge.  The core engine contains the standard, most widely used time series forecasting algorithms - linear and exponential regression, Holt Winters etc..

Learn more on the core statistical forecasting engine on this page.

R Logoforecasting engine


New from Version 11.0, Prophecy, offers tight / transparent integration with Microsoft R Open.  Microsoft R Open is an open source, optimised version of the R statistics and data science solution.

Users can run state-of-the art automatic sales forecasting algorithms, developed by renowned academic statisticians in R, from within Prophecy.

There are two alternative ways to forecast with R within Prophecy -  'Automatic' mode and 'R Expert' mode.R Forecasting templates supplied with Prophecy

'Automatic' mode requires no statistical or R expertise.  Prophecy takes care of everything, completely hides the complexity of R, and simply requires the user to select from one or more of the built-in R forecasting templates provided.

'R Expert' mode allows skilled R and statistical users to utilise R's full analytical, graphical and forecasting capabilities.  This mode splits 'Automatic' mode into two steps.  First, exporting data from Prophecy into R (RStudio).  Second, bringing it back into Prophecy once the forecasts have been generated in R / RStudio. The bit in the middle - generating the forecasts - is done by the user in RStudio.

The following sections explain in more detail:

'Automatic' forecasting mode

‘Automatic’ mode is designed to generate the most accurate forecasts and to be as easy to use as possible for non-statistical users.

The statistical models behind the scenes automate parameter selection and generate sensible forecasts with no user-intervention. The Prophecy user simply selects one or more of the special ‘R template’ files provided, specifies which products and customers to apply them to, and clicks ‘Run’.

Prophecy then runs R transparently in the background to generate the forecasts. The user can preview the resulting forecasts in a multi-dimensional grid view (similar to a Prophecy report) or view graphs showing the R forecast versus history and any existing forecast in Prophecy already.

The final stage in the ‘Automatic’ process is to apply the R forecasts back to the Prophecy database by clicking the 'Apply!' button. Alternatively, the forecasts that R has generated can be saved for later, and brought into Prophecy (if required) through the ‘R Expert’ mode import process.

View full movie on 

In summary, ‘Automatic’ mode is designed to require zero knowledge of R or statistics and to quickly get you as good a set of statistical forecasts as today’s state of the art time series forecasting allows.

'R Expert' forecasting mode

‘R Expert’ mode lets Prophecy users access the totality of R as a statistical and data science system.

99% of users will use Prophecy’s ‘Automatic’ mode and get great results, because the built-in automatic forecasting routines in R are comparable with the best commercial statistical forecasting engines.

Use ‘R Expert’ mode to export Prophecy sales history into R for analysis, transformation, graphing and forecasting. Once you’ve generated forecasts, use Prophecy’s ‘R Import’ button to read them back into Prophecy.

'Out of the box' forecasting templates

Prophecy comes with pre-built templates to use the following R forecast algorithms in both Automatic and Expert modes:

  • Arima 
    Returns best ARIMA model according to either AIC, AICc or BIC value. The function conducts a search over possible models within the default order constraints.
  • BATS
    Exponential smoothing state space model with Box-Cox transformation, ARMA errors, Trend and Seasonal components. As described in De Livera, Hyndman & Snyder (2011).
  • ETS
    Exponential smoothing state space model.
  • HoltWinters
    Parameters are determined by minimizing the squared prediction error.
  • Prophet
    Prophet is a procedure for forecasting time series data based on an additive model where non-linear trends are fit with yearly, weekly, and daily seasonality, plus holiday effects. It works best with time series that have strong seasonal effects and several seasons of historical data. Prophet is robust to missing data and shifts in the trend, and typically handles outliers well. Prophet is open source software released by Facebook's  Core Data Science team.
  • Multi-model Tournament
    Runs a tournament of methods (Arima, BATS, ETS, HoltWinters) and chooses the method with the lowest MAPE, as calculated by the accuracy() function of the forecast library.