| 摘要: | 本研究加入Hirshleifer et al.(2013)所提出之創新效率因子於三因子資產定價模型中，創新效率可同時描繪創新投入及創新產出，捕捉股票報酬之解釋力，藉以探討台灣電子工業上市櫃公司之創新效率對股票報酬之影響。研究採用追蹤資料進行分析，以單因子、三因子、以及四因子資產定價模型探討市場風險因子(〖MKT〗_t)、規模因子(〖SMB〗_t)、價值因子(〖HML〗_t)及創新效率因子(〖EMP〗_t)對電子工業上市櫃企業股票報酬於2010年7月至2021年6月期間之解釋能力。
實證結果顯示資本資產定價模型(Capital Asset Pricing Model, CAPM)模型之結果為顯著，台灣電子工業上市櫃企業市場組合月平均報酬率優於無風險資產月平均報酬率；而規模因子(〖SMB〗_t)、價值因子(〖HML〗_t)亦對股票橫斷面有顯著之解釋力。具體來說，小規模投資組合月平均報酬率會優於大規模投資組合月平均報酬率，而高帳面市值比因子投資組合月平均報酬率會優於低帳面市值比投資組合月平均報酬率，與Fama & French (1992)之發現相同。
而創新效率因子(〖EMP〗_t)亦對整體電子工業有一定之解釋力，與Hirshleifer et al. (2013)之發現相同，即創新效率高的股票相較於創新效率低的股票可獲得較高報酬；具體而言，高創新效率投資組合月平均年報酬率優於低創新效率投資組合月平均年報酬率為0.9642%。再進一步透過異質性分析發現創新效率因子(〖EMP〗_t)在光電業、通信網路業、電腦及周邊業較無解釋能力，於其他電子業在10%水準下顯著，電子通路業和電子零組件業在5%水準下顯著，而半導體產業在1%水準下顯著;說明了創新效果因應子產業之不同特性而有所迴異，由此建議基金經理人在選擇電子工業之標的和優化買賣時間時，充分考量創新活動對股票報酬的差異性影響。
;This study incorporates innovation efficiency factor proposed by Hirshleifer et al. (2013) into the three-factor asset pricing model to capture both input and output of innovation in boosting the explanatory power of how innovation efficiency affects the return of listed companies in the Taiwan electronics industry. Employing single-factor, three-factor, and four-factor asset pricing models, this study applies panel data analysis to investigate the effect of market risk factor (〖MKT〗_t) , size factor (〖SMB〗_t) , value factor (〖HML〗_t) , and innovation efficiency factor (〖EMP〗_t) on the stock returns of listed companies in the electronics industry from July 2010 to June 2021.
The empirical results show that the monthly average returns of a market portfolio comprised of listed companies in the electronics industry of Taiwan outperform those of risk-free assets, a finding that corroborates the Capital Asset Pricing Model (CAPM). Additionally, the size factor (SMB) and the value factor (〖HML〗_t) explain the cross-sectional stock returns well, where small-sized investment portfolios yield higher monthly average returns compared to large-sized investment portfolios, and investment portfolios with high book-to-market ratios deliver higher monthly average returns than those with lower book-to-market ratios, thus confirming the findings of Fama & French (1992).
Furthermore, innovation efficiency factor (〖EMP〗_t) also holds explanatory power for the overall electronics industry, consistent with the findings of Hirshleifer et al. (2013), suggesting that stocks with higher innovation efficiency tend to yield higher returns compared to those with lower innovation efficiency. Specifically, the high innovation efficiency investment portfolio′s monthly average annualized return outperforms the low innovation efficiency investment portfolio by 0.9642%. Further heterogeneity analysis reveals that the innovation efficiency factor (〖EMP〗_t) lacks explanatory power in the optoelectronics, telecommunications, and computer peripherals industries. However, it is significant at a 10% level in other electronic industries, at a 5% level in the electronic distribution and electronic components industry, and a 1% level in the semiconductor industry. This illustrates the varying degree of impact of innovation efficiency across sub-industries, thus calling for fund managers to thoroughly consider the differential impact of innovative activities on stock returns when selecting investment targets and optimizing the timing of buying and selling within the electronics industry. |
