I have been looking at moving-average forecasts and I'm a little confused regarding how useful this forecasting technique actually is.....

Lets say that I have I have the following data (for 12 months):

$\displaystyle 1000, 1140, 992, 843, 969, 979, 1106, 1029, 1152, 1083, 964, 974$

and using a three month average forecast I would have the following values:

$\displaystyle N/A, N/A, N/A, 1044.00, 991.67, 934.67, 930.33, 1018.00, 1038.00, 1095.67, 1088.00, 1066.33$

now lets say I had only 3 months worth of data here (1000, 1140, 992) and wanted to forecast the remaining months of the year.....the moving average forecast wouldn't really be suitable would it as I would end up with:

$\displaystyle 1044, 1058.67, 1031.56, 1044.74, 1044.99, 1040.43, 1043.39, 1042.93, 1042.25$

which would end up continuing to show values around 1042.

So would you not use the moving average forecast for anymore than perhaps a 1 or 2 month forecast beyond the data that you have? If so, why use the moving average forecast in the first place if you primarily use it for data you already have?

Thanks