Estimation

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Estimation
Message posted by Maria on September 27, 2000 at 12:00 AM (ET)

I asked this before but I didn't give enough information. I will appreciate your help.
Lets say that a company is trying to introduce a new brand and uses two methods to find out which one is best to use. One of the methods used is by obtaining actual sales data for one month and two after introduction of a new item, and to apply the percentage growth rate as a forecast. The other method used a model based on product life cycle ideas. As a Test, historical data of 38 previous new product introductions were used to forecast deamnd using both methods. What method seems consistently too optimistic or too pessimistic, Which method seems to be more accurate most of the time and any indication that one method or the other is liable to give widly poor estimates?
Method 1 mean:-3027; median:-1681; trmean -2614; stdev 4105; semean 666; min -18566; max 3223; Q1 -4021; Q3 -765.
Method 2: mean 278; median 352 trmean 373; stdev 2028; semean 329; min -6466; max 4478; Q1 -672 Q3 1185.

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Estimation
Message posted by JG on September 28, 2000 at 12:00 AM (ET)

Every forecasts or estimate of what happens next
involves a 'quess' about how the future and past are related to
each other. 'Standard' forecasting techniques are very different
from product life cycle ideas and there are many other ways of
solving the problem.
I would assume that 'appropriate' life cycle models should be much more accurate in your case that 'standard' forecasting techniques, if you are using the appropriate life cycle for that particular product.
Thus, for any given product you should have an appropriate life cycle model and you should update these models periodically, etc. An example of how things change is the arab oil embargo, OPEC, etc. Suddenly all forecasts of oil supplies and oil prices became obsolete
because the problem became one of political and not economic or technological. Using you terminology, the life cycle changed drastically.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.