本文蒐集台灣2004年至2020年間共215家上市上櫃製造業公司追蹤資料，藉由Cobb-Douglas生產函數，採用靜態與動態追蹤資料模型探討研發對產出之影響。靜態模型結果顯示，在 ICT產業中不論其次產業為何，研發費用對於營業利益皆有顯著的正向貢獻。動態模型實證結果顯示，研發之產出效果皆較靜態模型低。換言之，若忽略前期產出對當期產出之影響將高估研發之產出效果。本文亦探討ICT與非ICT行業中，其研發的產出效果是否有所差異，結果顯示ICT業之研發的產出效果大於非ICT行業。最後探討研發的產出效果是否隨公司規模不同而有所差異，結果顯示大型廠商研發的產出效果大於中小型廠商研發的產出效果。

We use the panel data of Taiwan's manufacturing firms from 2004 to 2020 to study the output effect of R&D and factor inputs. Based on the Cobb-Douglas production function, static and dynamic panel data models are used to explore the output effects of R&D. We find that R&D positively affects profits in the ICT industry and its subsectors regardless of whether a static or dynamic model is applied.; however, the magnitude of the output effect is smaller when using a dynamic model than when using a static model. Neglecting the lagged output effect in the model causes an overestimation of R&D's output effect. We also investigate whether R&D's output effects differ between ICT and non-ICT industries and find that the effect is larger in the ICT industry than in the non-ICT industry. Finally, we examine whether R&D's output effect changes with firm size and find that R&D's output effect on small and medium-sized firms is smaller than that of large firms.

論文審定書i
摘要ii
Abstractiii
目錄iv
圖次vi
表次vii
第一章 緒論1
第二章 文獻回顧3
第三章 實證方法7
第一節 實證模型設定7
第二節 靜態追蹤資料模型之選用8
2.1 混合最小平方法(Pooling OLS)8
2.2 固定效果模型(Fixed Effect Model)8
2.3 隨機效果模型(Random Effect Model)11
2.4 豪斯曼檢定(Hausman Test)13
2.5 小結14
第三節 動態追蹤資料模型之選用14
3.1 Anderson & Hsiao之工具變數法(IV)15
3.2 Arellano & Bond之差分廣義動差法(difference GMM)16
3.3 Blundell & Bond之系統廣義動差法(system GMM)16
3.4 過度認定檢定與二階自我相關檢定(Hansen Test & AR(2) Test)17
3.5 小結17
第四節 追蹤資料單根檢定 (Panel Unit Root Test)18
4.1 Levin-Lin-Chu(LLC)檢定19
4.2 Im-Pesaran-Shin(IPS)檢定19
第四章 實證結果21
第一節 資料來源與敘述統計21
第二節 基礎模型之實證結果27
2.1 靜態追蹤資料模型之實證結果27
2.2 動態追蹤資料模型之實證結果30
第三節 總體衝擊事件之實證結果32
3.1 2008年金融海嘯(Financial Crisis of 2008)32
3.2 2012年歐債危機(European Debt Crisis)33
3.3 電價調漲34
3.4 基本工資調漲35
第四節 產業與規模效應之實證結果40
第五章 結論43
參考文獻44
附錄47

一、中文部分
1. 吳俊諺(2006)，「研究發展對展出與市值之影響─以半導體產業為例」，國立交通大學管理科學系碩士論文。
2. 陳旭昇(2013)，「時間序列分析 : 總體經濟與財務金融之應用」，雙葉書廊，台北市。
3. 陳宗群(2017)，「知識型企業知識轉移對廠商經營績效的影響 : 以台灣半導體產業為例」，《華人經濟研究》，15(2)，47 – 59。
4. 陳強(2014)，「高級計量經濟學及Stata應用」，高等教育出版社，北京市。
5. 莊奕琦、許碧峰(1999)，「研究發展對生產力的貢獻及產業間的外溢效果：台灣製 造業實證」，《經濟論文》，27(3)，407 – 432。
6. 楊志海、陳忠榮(2002)，「研究發展，專利與生產力—台灣製造業的實證研究」， 《經濟論文叢刊》，30(1)，27 – 48。
7. 蔡慧真(2006) ，「廠商研發能力對績效影響之研究」，國立臺北大學企業管理學系博士論文。
8. 蔡斯芳(2017)，「臺灣製造業研發投入對收益的影響分析」，國立高雄應用科技大學國際企業系碩士論文。
9. 歐陽利姝(2008)，「研發、研發外溢與員工學歷結構差異對台灣資訊電子業研
發廠商生產力之貢獻」，《經濟論文叢刊》，36(4) ，515 – 550。

二、英文部分
1. Arellano, M., and S. Bond(1991), “Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations.”, Review of Economics Studies, 58, 277 – 297.
2. Anderson, T., and C. Hsiao(1981), “Estimation of Dynamic Models with Error Components.”, Journal of the American Statistical Association, 76, 598 – 606.
3. Baltagi(2005), Econometric Analysis of Panel Data, John Wiley & Sons Inc., New York.
4. Blundell and Bond(1998), “Initial conditions and moment restrictions in dynamic panel data models”, Journal of Econometrics, 87, 115 – 143.
5. Cuneo, P., and J. Mairesse(1984), “Output and R&D at the Firm Level in French Manufacturing”, R&D, Patents and Output, 375 – 392, University of Chicago Press.
6. Cohen(1995), “Empirical Studies of Innovative Activity”, In: P. Stoneman, Ed., Handbook of the Economics of Innovation and Technological Change, 182 – 264.
7. Granger, C., and P. Newbold(1974), “Spurious Regressions in Econometrics.”, Journal of Econometrics, 2, 111 – 120.
8. Greene(2012), Econometric Analysis (7th Edition), Pearson, London.
9. Griliches(1980), “Returns to Research and Development Expenditures in the Private Sector”, New Development in Productivity Measurement and Analysis, University of Chicago Press, 419 – 461.
10. Griliches, Z. and J. Mairesse(1983), “Comparing Output Growth: an exploration of French and U.S. industrial and firm data”, European Economic Review, 21, 89 – 119.
11. Griliches, Z. and J. Mairesse(1984), “Output and R&D at the Firm Level”, R & D, Patents, and Productivity, 339 – 374, University of Chicago Press.
12. Griliches(1986), “Productivity, R&D, and Basic Research at the Firm Level in the 1970's ”, The American Economic Review, 76(1), 141 – 154.
13. Hosung(2015), “An Alternative System GMM Estimation in Dynamic Panel Models”, Journal of Economic Theory and Econometrics, 26(2), 57 – 78.
14. Legge(2000), “Around the industry”, Employee Benefit Plan Review, 55(1), 64
15. Liu(2019), “R&D Expenditure Spillovers and Productivity Empirical Study of Taiwanese Optoelectronic Industry”, Asian Economic and Financial Review, 9(8), 888 – 900.
16. Lokshin(2008), “The Productivity Effects of Internal and External R&D: Evidence from a Dynamic Panel Data Model”, Oxford Bulletin Of Economics And Statistics, 70(3), 399 – 413.
17. Romilio Labra(2018), “Estimating dynamic Panel data. A practical approach to perform long panels”, Revista Colombiana de Estadística, 41(1), 31 – 52.
18. Roodman(2009), “How to do xtabond2: An introduction to difference and system GMM in Stata”, The Stata Journal, 9, 86 – 136.
19. Sher, P. J. and Yang, P. Y.(2005), “The effects of innovative capabilities and R&D clustering on firm performance: the evidence of Taiwan’s semiconductor industry”, Technovation, 25(1), 33 – 43.
20. Tarek Sadraoui(2009), “A Dynamic Panel Data Analysis For R&D Cooperation and Economic Growth”, International Journal of Foresight and Innovation Policy, 5(4), 218 – 233.
21. Tsai, K. H. and Wang, J. H.(2004), “The R&D performance in Taiwan’s electronics industry: a longitudinal examination”, R&D Management, 34(2), 179 – 190.
22. Tsai (2005), “R&D output and firm size: a nonlinear examination”, Technovation, 25(7), 795 – 803.
23. Wakelin K.(2001), “Productivity growth and R&D expenditure in UK manufacturing firms”, Research Policy, 30(7), 1079 – 1090.
24. Zeila Serrasqueiro(2009), “Growth and Profit in Portuguese Companies : A Dynamic Panel Data Approach”, Economic Interferences, 11(26), 565 – 573.