I'd like to use Simple Exponential smoothing to forecast the lead-time demand for inventory control, I have monthly data and LT+1 is equal to 5 months, can I do a forecast using SES which gives me a one-period-ahead forecast and then multiply it by 5 to get the forecast for LT+1 ? what is the best solution to get h step ahead forecast(like LT+1)when the forecast method is SES? Thank you
- Anybody can ask a question
- Anybody can answer
- The best answers are voted up and rise to the top
Yes, for Single Exponential Smoothing, the lead time demand (which is the sum of the forecasts over the lead time) is necessarily a simple multiple of a one-step-forecast.
If you use a more complex model (with seasonality and/or trend), this will of course not hold any more. In addition, you will need to think about safety stocks and quantile forecasts.
I recommend this free online textbook on forecasting.