以多元迴歸實證分析專利價值與專利因素

Empirical Study on Patent Indicators and Patent Value by Multiple Regression Analysis

侯門(Hou, Men)

指導教授 ： 范建得

國立清華大學/科技管理學院/科技法律研究所/碩士(2015年)

https://doi.org/10.6843/NTHU.2015.00530

摘要

世界各國對於專利交易情形之發展已日漸頻繁，因此自有其必要探討、分析專利價值與各專利因素之間之相關性，以利於建構專利價值模型。本研究之對象係為工業技術研究院所有之專利，以分析其專利價值與各專利因素間之相關性。

　　本研究先整理先前文獻關於專利價值與各專利因素之相關性之探討及專利因素影響專利價值之理論，以確立本研究所欲分析之專利因素及欲使用之研究方法。在研究方法上，本研究使用先前研究所使用之迴歸分析，或更進一步言之，主要以對數迴歸分析為主。本研究先利用單一迴歸分析找出與專利價值具有顯著相關性之顯著專利因素，再基於該等顯著專利因素以多元迴歸分析之。透過前述分析可發現本研究選定之專利因素在單一迴歸及多元迴歸上之分析結果，得以實證先前文獻之理論，並額外發現先前文獻無法證明顯著性之部份專利因素之顯著性。

　　由於本研究之研究方法之先天限制，本研究僅能得出在迴歸模型中專利價值與專利因素之相關性而非因果關係，因此專利價值與專利因素間之因果關係，仍有待未來研究闡述，以連同本研究之成果成為未來制定專利價值模型時之參考依據。

關鍵字

專利 ； 價值 ； 發光二極體 ； 多元迴歸 ； 實證分析 ； 工業技術研究院

並列摘要

Patent transaction is becoming frequent through whole world, hence it’s necessary to study and analyze the relation among patent value and various patent indicators to develop patent value model. The subjects of this article are patents owned by Industrial Technology Research Institute (ITRI), these patents are analyzed for statistical relation among their patent value and patent indicators.

This article reviews prior references discussing on relations among patent value and patent indicators and theories of how patent indicators affect patent value, to establish analytical method of this research and choose patent indicators about to be analyzed by this research. About the analytical method, this article is conducted on regression analysis used by prior research, more specifically, on logarithmic regression analysis, essentially. At first, this article use simple regression analysis to retrieve those indicators with significant relation with patent value then apply multivariate regression analysis on those indicators. By this method, this article reveals the result of chosen indicators at simple regression level and at multivariate regression level to empirically prove theories from prior references. Additionally, this article discovers the significance of some indicator found as being insignificant in prior references.

Due to natural restrain on the analytical method used in this article, this article can only establish the relation of patent value with patent indicators rather than the causality thereof. Such causality shall only be investigated in future for being references for designing patent value model with the result contributed by this article.

並列關鍵字

Patent ； Value ； Light Emitting Diode ； Multiple Regression ； Empirical Study ； Industrial Technology Research Institute

參考文獻

譚克平：〈極端值判斷方法簡介〉，《台東大學教育學報》第19卷第1期〔2008年6月〕，頁131-150。

吳耀宏：〈分析美國化工產業的專利範疇、RTA、平均專利年齡對市值之影響〉（台北：國立臺北大學企業管理學系碩士論文，2013）。取自臺灣博碩士論文知識加值系統。

Abrams, D. S., Akcigit, U., and Popadak, J. (2013). Patent Value and Citations: Creative Destruction or Strategic Disruption? (No. w19647). National Bureau of Economic Research.

Alan Pilkington, Linda L. Lee, Casey K. Chan and Seeram Ramakrishna. (2009). Defining key inventors: A comparison of fuel cell and nanotechnology industries. Technological Forecasting and Social Change, 76(1), 118-127.

Alfonso Gambardella, Dietmar Harhoff and Bart Verspagen. (2008). The value of European patents. European Management Review, 5(2), 69-84.