Predictive accuracy of survey-based forecasts for numbers of filings at the European Patent Office
This study examines the predictive accuracy of the forecasts for patent filings from a sequence of annual surveys of patent applicants at the European Patent Office, particularly regarding the extent to which applicants could foresee the effects of the 2008/2009 recession. It also investigates the p...
Saved in:
Published in | World patent information Vol. 35; no. 3; pp. 187 - 200 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
Elsevier Ltd
01.09.2013
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | This study examines the predictive accuracy of the forecasts for patent filings from a sequence of annual surveys of patent applicants at the European Patent Office, particularly regarding the extent to which applicants could foresee the effects of the 2008/2009 recession. It also investigates the possibilities for new methods to calculate forecasts for survey results.
Applicants were selected randomly each year, and opinions were sought about numbers of filings in the previous year, the current year and the following two years. It is found that two and three year predictions were better for worldwide first filings than for EPO Total filings (sum of European direct and PCT International Phase filings, excluding divisional filings), and differences are established between main blocs of residence of the applicants in this regard. An investigation is made to examine the possible benefits of indirect transfer forecasts for total EPO filings via worldwide first filings, and of indirect forecasts of PCT regional phase entries from PCT international phase filings. Although the recession was not well anticipated by the applicants, it was slightly better anticipated in terms of worldwide first filings and PCT regional phase entries, than for total EPO filings. A combination approach may be appropriate for future surveys.
Suggestions are made for the further development of analysis methods that can be tried out on future surveys.
•MAPE values suggest a reasonable survey forecasting accuracy for patent filings.•2 and 3 year predictions for first filings are better than for Total EPO filings.•The recession was anticipated badly, worse for total EPO filings than first filings.•Total EPO filings can be forecasted indirectly from worldwide first filings forecasts.•PCT regional phase entries can similarly be forecasted from international phase. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0172-2190 1874-690X |
DOI: | 10.1016/j.wpi.2013.04.006 |