ExcelBanter - forecast production

There's no simple answer. The method you choose should depend on things like
how many prior years of production you have available for a baseline, what
that baseline looks like (generally straight-line? trended up or down?
cyclical/seasonal? etc.), your subjective knowledge of any changes the
company has in mind for its production facilities, and so on.

There are three basic approaches to forecasting, and each is available in
Excel:

1. Regression. Excel supports this with functions like TREND (much better
than FORECAST, IMHO, because TREND accepts multiple predictors), LINEST, and
so on. There's a Regression tool in the Analysis Toolpak, and it's helpful
if you're unfamiliar with Excel's regression functions.

2. Moving averages. Again, you can roll your own formulas for moving
averages or rely on the Analysis Toolpak.

3. Smoothing, closely related to moving averages. Again, the Analysis
Toolpak has a smoothing tool.

ARIMA analysis (not readily available in Excel, which does not directly
calculate ACFs and PACFs) gives you objective information that helps you
decide which approach, or which combination of approaches, to use. Without
ARIMA analysis, you have to rely on experience, trial-and-error, or a
dartboard to choose. That said, you usually need a baseline of roughly 50
periods to begin ARIMA identification. It takes that many to reliably model
the structure of the correlated error in the baseline. There are various
statistical packages, some of them freeware, that offer ARIMA analysis.

--
C^2
Conrad Carlberg

Excel Sales Forecasting for Dummies, Wiley, 2005

"jrM..." wrote in message
...
I would like to be able to forecast next years production using previous
years production figures.