Activity Number:
|
356
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Business & Economics Statistics Section*
|
Abstract - #301183 |
Title:
|
Accuracy of Nonlinear Forecast Models Based on a Set of Historical Values Using Fixed Starting Points
|
Author(s):
|
Martin Selzer*+
|
Affiliation(s):
|
National Exchange Carrier Association
|
Address:
|
80 South Jefferson Road, Whippany, New Jersey, 07981, USA
|
Keywords:
|
Forecasting ; Non-linear Regression ; Data Updates ; Visual Basic Applications
|
Abstract:
|
A forecasting model can be framed such that the starting point over a forecast horizon is known and historical values are available to develop projections. This kind of model is useful for forecasting problems where a value is observed in an initial period and then updated in future periods. The goal is to predict the last period updated value from the initial observed value. The current research uses data from the National Exchange Carrier Association earnings evaluation process to test non-linear models to predict revenues reported by firms.
|