Imperial College London
Portrait Picture

Emeritus ProfessorNigelMeade

Business School

Emeritus Professor of Quantitative Finance
 
 
 
//

Contact

 

n.meade

 
 
//

Location

 

53 Prince's GateSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@inbook{Meade:2006:10.1007/3-540-35992-3_4,
author = {Meade, N},
booktitle = {Forecasting Innovations: Methods for Predicting Numbers of Patent Filings},
doi = {10.1007/3-540-35992-3_4},
pages = {41--72},
title = {An assessment of the comparative accuracy of time series forecasts of patent filings: The benefits of disaggregation in space or time},
url = {http://dx.doi.org/10.1007/3-540-35992-3_4},
year = {2006}
}
Download

RIS format (EndNote, RefMan)

TY  - CHAP
AB  - This work is concerned with methods for forecasting the filing of patents and was earned out in conjunction with the European Patent Office. The filings data were subdivided by:. Blocs-European Patent Convention countries, Japan, US and the Rest of the World. Industries-main Fields of Technology according to headings A-H of the International Patent Classification (WIPO 2000). The issues addressed are: the benefits of multivariate models versus univariate ones in exploiting any correlations between the filings in different blocs or industries; the effect of aggregation over time (from monthly to annual data) and the effect of aggregation by bloc or by industry on the accuracy of the forecast of total EPO filings. Two approaches are used: the ARIMA framework and the dynamic linear model (DLM) in both univariate and multivariate modes. The main results are: monthly data does tend to provide greater accuracy in annual forecasts; there are no significant benefits to be gained by multivariate modelling and no significant benefits are found from aggregating over blocs or industries. There are benefits from using monthly data, rather than annual data. The best modelling approaches are, for monthly data, the univariate dynamic linear model; for annual data either the univariate ARIMA or DLM could be used. The recommended forecast-ing approach provides a benchmark against which other forecasts drawing on different data sources can be compared. The filing of an application for a patent is the first step in achieving protection for intellectual property. The three major patent offices are the European (EPO), the Japanese (JPO) and the US (USPTO) offices. The examination of each application is a labour intensive process, involving technological and legal expertise. The motivation for this study of the forecasting of patent filings with the EPO is that the forecasts are a prerequisite for manpower planning at an aggregate scale and at the level of availability of expertise in dif
AU  - Meade,N
DO  - 10.1007/3-540-35992-3_4
EP  - 72
PY  - 2006///
SN  - 9783540359913
SP  - 41
TI  - An assessment of the comparative accuracy of time series forecasts of patent filings: The benefits of disaggregation in space or time
T1  - Forecasting Innovations: Methods for Predicting Numbers of Patent Filings
UR  - http://dx.doi.org/10.1007/3-540-35992-3_4
ER  -
Download