Replication Data for: Exponential Smoothing Forecasts: Taming the Bullwhip Effect when Demand is Seasonal.

This dataset comprises of all the simulated time-series used in the paper: Exponential Smoothing Forecasts: Taming the Bullwhip Effect when Demand is Seasonal.

The data is as follows. 14 different csv files are presented. Each file contains 20 time-series, of 6000 periods each, that are generated using the same parameters.

As described in the paper, the structure of the time-series in each of the files varies in the following parameters:
* The underlying distribution (a normally distributed i.i.d process, or an ARIMA (0,1,1) process).
* The seasonality of the time-series (no seasonality, seasonality of periodicity 49, 50, or 51).
* For the normally distributed, seasonal demands, the strength of the seasonality (weak, medium, strong seasonality).