DSpace community: 財務金融學系碩士在職專班

後現貨 ETF 時代下比特幣期現貨高頻基差動態：制度性時間摩擦、流動性拔河與不對稱尾部風險分析;後現貨 ETF 時代下比特幣期現貨高頻基差動態：制度性時間摩擦、流動性拔河與不對稱尾部風險分析

Tue, 20 Oct 2026 08:00:18 GMT

title: 後現貨 ETF 時代下比特幣期現貨高頻基差動態：制度性時間摩擦、流動性拔河與不對稱尾部風險分析;後現貨 ETF 時代下比特幣期現貨高頻基差動態：制度性時間摩擦、流動性拔河與不對稱尾部風險分析 abstract: 本文利用高頻1分鐘價格資料，探討加密貨幣期貨市場（CME）與現貨市場（Binance）之間的絕對基差（absolute basis）的動態，檢視日內價格在不同時段的週期性變化，跨市場之間的流動性拔河，以及極端市場風險衝擊下之非對稱邊際疊加效應。本文以芝加哥交易所CME期貨價格跟幣安Binance的現貨價格為樣本資料，樣本期間為2024年1月1日至2025年7月25日，共計350，917 筆1分鐘資料觀測值。實證模型考慮時間周期性與固定效應，實證結果顯示跨市場的「制度性時間摩擦」確實對24小時不間斷之加密貨幣定價過程造成顯著的摩擦成本。CME 週末休市引發之「週五效應」，以及每日結算真空期（HOUR21）所導致之「定價致盲」，皆促使造市商為規避資訊不對稱與無法對沖之風險而採取防禦性撤單，導致絕對基差擴大。另外，當 Binance 現貨交易量大於 CME的期貨交易量時，絕對基差呈現負向的收斂，證實BTC現貨市場主導跨市場之價格發現，與具有流動性優勢。本研究發現在極端風險衝擊下，基差呈現「左尾恐慌擴張、右尾狂熱收斂」之不對稱階梯式特徵。當市場面臨極端下跌（前10%、5%、1%閃崩）時，隨著跌幅加深，造市商因恐懼單邊存貨爆倉而引發流動性不對稱枯竭，導致基差呈階梯式擴大；相反地，當價格極端暴漲時，加密貨幣投資人特有之錯失恐懼症（FOMO）引起單邊的積極追價之買單，龐大流動性反而強勢地將期現貨價差強行抹平。綜上所述，本文驗證時間制度差異對資產定價的影響，發現加密貨幣具有「跌越深、摩擦越大；漲越狂、價差越小」之微結構特性。本文的研究結果，對於高頻套利者之風險控管，造市商之動態存貨決策，以及金融監管機構（如 SEC和CFTC）對於跨市場之系統性風險控管方面，皆能提供具體實務操作與政策監管上的參考。關鍵詞：加密貨幣；絕對基差；時間摩擦；流動性拔河；聯合尾部估計；不對稱撤單；錯失恐懼(FOMO)效應;This dissertation utilizes high-frequency one-minute data to comprehensively investigate the microstructural dynamics of the absolute basis between the regulated traditional futures market (Chicago Mercantile Exchange， CME) and the cryptocurrencies spot market (Binance). We systematically evaluate the impacts of institutional temporal frictions， cross-market liquidity tug-of-war， and asymmetric marginal overlapping effects triggered by extreme market risk shocks. The data include Bitcoin futures prices from the CME and spot prices from Binance， spanning from January 1， 2024， to July 25， 2025. Based on the final effective sample of 350，917 observations， we find that cross-market “institutional temporal frictions” introduce significant friction costs into the 24/7 continuous pricing mechanism of cryptocurrencies. The “Friday effect” induced by CME weekend closures， alongside the “pricing blindness” caused by the daily settlement vacuum period (HOUR 21)， compels market makers to execute defensive order cancellations to hedge against information asymmetry and unhedgeable risks， thereby significantly widening the absolute basis. Next， within the regular liquidity tug-of-war， when Binance spot trading volume expands significantly relative to the CME， the absolute basis exhibits a highly statistically significant negative convergence. This confirms that the crypto-native spot market dominates cross-market price discovery and maintains a liquidity premium under normal market conditions. Third， a major novel finding of this study is that under extreme risk shocks， the basis exhibits a pronounced asymmetric， stepwise overlapping feature characterized by “left-tail panic expansion and right-tail euphoria convergence.” When the market encounters extreme downturns (the top 10%， 5%， and 1% flash crashes)， the absolute basis widens in a stepwise fashion as the magnitude of the decline intensifies. This expansion occurs because market makers， fearing unilateral inventory liquidation， trigger asymmetric liquidity depletion. Conversely， during extreme surges， the unique fear-of-missing-out (FOMO) effect inherent in cryptocurrency markets drives aggressive， unilateral price-chasing buy orders， generating massive liquidity that forcefully compresses and erases the futures-spot price discrepancy. Overall， the empirical results confirm the distortions in asset pricing caused by institutional temporal discrepancies， mapping out a unique microstructural blueprint for cryptocurrencies where deeper price drops yield greater frictions， whereas wilder surges result in smaller bases. These findings offer concrete references for both practical implementation and policy oversight， particularly for risk management among high-frequency arbitragers， dynamic inventory pricing for market makers， and financial regulatory authorities (such as the SEC and CFTC) in mitigating cross-market systemic risks. Keywords: Cryptocurrency; absolute Basis; temporal frictions; liquidity tug-of-war; joint tail estimation; asymmetric order cancellation; FOMO (fear of missing out) effect

投資人情緒指標與台股指數報酬率之關係;Investor Sentiment and Stock Index Return: Evidence from the Taiwan Stock Market

Tue, 20 Oct 2026 08:00:14 GMT

title: 投資人情緒指標與台股指數報酬率之關係;Investor Sentiment and Stock Index Return: Evidence from the Taiwan Stock Market abstract: 本論文探討整合型情緒指標對台股報酬的預測能力，並分析投資人情緒相較於傳統總體經濟指標的預測表現。本文以台灣加權股價指數超額報酬率為預測標的，樣本期間為2004年5月至2025年12月，評估情緒與總經指標之樣本內預測與樣本外預測之表現，比較等權重加權平均法 (EW)、主成分分析法 (PCA) 與偏最小平方法 (PLS) 之預測能力。實證結果顯示：第一，在樣本內預測表現方面，投資人情緒指標對股票報酬的預測力顯著優於傳統總體經濟指標；第二，樣本內預測結果顯示 PLS 方法的預測能力優於 PCA 與 EW 方法；第三，以均方差 (root mean squared error， RMSE) 評估樣本外預測表現之結果顯示，PLS 所建構的情緒指標並未優於 PCA 與 EW 方法。;This thesis evaluates the predictive ability of aggregate sentiment index on excess returns of Taiwan stock index， and compares the forecasting performance of investor sentiment relative to traditional macroeconomic fundamental indicators. Utilizing the excess returns of the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) as the forecast target， covering the period from May 2004 to December 2025， we evaluate in-sample and out-of-sample forecast performances of investor sentiment and macroeconomic indicators， among approaches of equal-weighted average (EW)， principal component analysis (PCA)， and partial least squares (PLS). The empirical findings are three-fold: First， in terms of in-sample forecast performance， the aligned investor sentiment index exhibits significantly outperforms the aligned index of macroeconomic fundamental indicators. Second， regarding the in-sample forecast performance of the PLS method is better than those of PCA and EW methods. Third， the out-of-sample forecast evaluation based on the root mean squared error (RMSE) shows the aligned investor sentiment constructed via PLS does not outperform both PCA and EW approaches.

綠建築標章分級對住宅房價溢價幅度影響之研究：以台北市為例;Impact of Green Building Certification Levels on Residential Price Premiums — A Case Study of Taipei City

Tue, 20 Oct 2026 08:00:02 GMT

title: 綠建築標章分級對住宅房價溢價幅度影響之研究：以台北市為例;Impact of Green Building Certification Levels on Residential Price Premiums — A Case Study of Taipei City abstract: 近年來，氣候變遷與淨零排放議題日益受到重視，綠建築標章作為衡量建築物節能與環境效益之重要指標，其經濟價值是否確實反映於房價中，成為不動產資產定價研究中值得關注之議題。然而，現有文獻對於不同標章等級之溢價是否呈現遞增關係、以及建商品質是否混淆標章與房價之關聯，實證證據仍相對有限。本研究首次於台灣綠建築溢價研究中導入建商固定效果模型，以台北市2011年至2026年共286個已取得EEWH綠建築標章之住宅社區、16，612筆實價登錄交易資料為樣本，採用特徵價格法估計銅級、銀級、黃金級（含）以上三個標章等級相對於合格級之房價溢價幅度，並檢視此效果於精華區與非精華區、疫情前後及預售屋與成屋市場間之異質性。實證結果顯示，未控制建商特性時，銀級與黃金級以上標章相對於合格級呈現負向係數；納入建商固定效果後，係數方向完全翻轉，銅級（+12.01%）與銀級（+3.89%）均呈顯著正向溢價，黃金級（含）以上則未達顯著，顯示建商品質為關鍵混淆因子。異質性分析發現，成屋市場之銅級溢價（+23.29%）遠高於預售市場（−10.33%），呈現顯著資訊不對稱效應；黃金級（含）以上於預售市場亦呈顯著負向（−6.36%），標章溢價並非隨等級單調遞增。本研究並將溢價換算為實際金額，發現銅級標章每坪可帶來約9.81萬元之經濟價值，有助於釐清綠建築標章之資本化程度，並為綠色金融政策提供實證依據。;As climate change and net-zero emissions draw growing global attention， green building certification has become an important indicator of a building′s energy efficiency and environmental performance. Whether this certification is capitalized into housing prices remains an open question in real estate asset pricing research， with limited evidence on whether premiums rise monotonically across certification levels and whether developer quality confounds the certification-price relationship. This study uses a sample of 16，612 real price registration transactions from 286 EEWH-certified residential communities in Taipei City over the period 2011 to 2026. Employing a hedonic pricing model with developer fixed effects， we estimate the housing price premiums of the Bronze， Silver， and Gold-and-above (Gold and Diamond levels combined， given the limited number of Diamond-certified communities) certification levels relative to the Qualified level， and further examine the heterogeneity of these premium effects across prime and non-prime districts， before and after the COVID-19 pandemic， and between presale and existing-home markets. The empirical results reveal that， without controlling for developer characteristics， the Silver and Gold-and-above certifications exhibit negative coefficients relative to the Qualified level. However， after incorporating developer fixed effects， the signs of the certification coefficients reverse entirely: both Bronze (+12.01%) and Silver (+3.89%) certifications show significant positive premiums， while the Gold-and-above certification remains statistically insignificant， indicating developer quality as a critical confounding factor. Heterogeneity analyses further reveal that the Bronze premium in the existing-home market (+23.29%) substantially exceeds that in the presale market (−10.33%)， suggesting a pronounced information asymmetry effect; the Gold-and-above certification also shows a significant negative premium in the presale market (−6.36%)， indicating a non-monotonic premium pattern across certification levels. Converting percentage premiums into monetary terms， this study finds that Bronze certification yields an economic value of approximately NT$98，100 per ping (3.3 m²). We further discuss developers′ potential economic incentives for pursuing green building certification through the lens of floor area ratio bonus policies. To the best of our knowledge， this is the first study in Taiwan to incorporate developer fixed effects into the estimation of green certification premiums in the residential market. The findings help clarify the degree of capitalization of green building certification in Taipei′s housing market， provide a baseline reference for the mandatory disclosure policy on real estate sustainability ratings scheduled to take effect in 2026， and offer empirical implications for the advancement of green finance policy in Taiwan.

轉機股是否有超額報酬之實證研究;轉機股是否有超額報酬之實證研究

Tue, 20 Oct 2026 07:59:26 GMT

title: 轉機股是否有超額報酬之實證研究;轉機股是否有超額報酬之實證研究 abstract: 本研究旨在探討台灣股票市場中，長期陷入經營困境後成功扭轉局勢、由虧轉盈之「轉機股」是否具備超額投資報酬，並進一步剖析影響異常報酬之關鍵因素。本研究以台灣上市櫃全產業（排除金融業）公司為篩選標的，量化定義轉機股為「連續四個年度每股盈餘（EPS）小於零，且第五個年度達到損益兩平或開始獲利」之企業，由2006年起到2024年，建立19個年度投資組合；為了避免倖存者偏誤（Survivorship Bias），同時篩選TEJ已下市櫃資料庫中，在研究周期內仍可以在市場正常買賣的公司，上市櫃以及已下市櫃公司共計篩選出396個轉機股。實證方法採用事件研究法，計算虧轉盈資訊公告前後的累積異常報酬（CAR）與買進持有異常報酬（BHAR），並建立三組情境稍有不同的迴歸模型檢定財務困境嚴重程度（Z-score）、自由資產、股東權益報酬率（ROE）、股價淨值比、公司規模、董監持股比例以及股權集中度變動等變數對超額報酬之影響。研究結果顯示，台灣股市確實存在跨越「EPS = 0」的門檻效應，轉機股在盈餘資訊公告月後 3 個月內（研究窗期 (1，3)）以及公告月前2個月到公告月（研究窗期 (-2，0)）存在顯著大於零的超額報酬，有明顯的市場過度反應效應與盈餘慣性現象（PEAD），PEAD現象約莫就三個月左右。若未經進一步策略選擇轉機股而長期持有至一年半或兩年，投資組合將面臨大幅度累積虧損之風險。在關鍵影響因素方面，實證指出低股價淨值比與小公司規模是延長轉機股獲利天期與提升超額報酬的核心基本面因子，比較特別的是，財務困境嚴重程度Z-score的反向影響相當明顯，這三個變量的顯著性、趨勢性與穩定性在三組迴歸分析結果的交叉比對中都較好，此實證結果顯示，Z-score較低財務風險稍高、股淨比較低估值較低且規模較小的企業，投資報酬率較高。本研究發現，轉機股的超額報酬主要受到風險溢價與價值低估的共同驅動，而透過股價淨值比、公司規模以及Z-score幾個要素的篩選，確實可以提高投資勝率與報酬率。;This study investigates whether "turnaround stocks"—defined as firms transitioning from chronic financial distress to profitability—generate abnormal investment returns in the Taiwan stock market， and analyzes the key determinants influencing these excess returns. Utilizing an empirical sample of listed， OTC， and delisted non-financial firms from 2006 to 2024， we construct 19 annual portfolios. Turnaround stocks are quantitatively designated as firms experiencing negative earnings per share (EPS) for four consecutive fiscal years followed by a breakeven or profitable fifth year. Delisted firms from the TEJ database are explicitly incorporated to eliminate survivorship bias， yielding a total sample of 396 turnaround stocks. Methodologically， an event study approach is deployed to compute cumulative abnormal returns (CAR) and buy-and-hold abnormal returns (BHAR) surrounding earnings reversal announcements. Three econometric regression specifications with distinct situational scenarios are further constructed to examine the impacts of financial distress severity (Altman′s Z-score)， free assets， ROE， price-to-book (P/B) ratio， firm size， insider ownership， and ownership concentration changes on these abnormal returns. The empirical findings reveal a distinct threshold effect surrounding the "EPS = 0" boundary in the Taiwanese equity market. Turnaround stocks exhibit statistically significant positive abnormal returns in event windows (1， 3) and (-2， 0)， providing clear evidence of a market overreaction effect and a post-earnings-announcement drift (PEAD) phenomenon that persists for approximately three months. Conversely， holding these turnaround portfolios long-term for 1.5 to 2 years without strategic fundamental screening exposes investors to severe cumulative losses. Regarding key determinants， a low P/B ratio and small firm size serve as core fundamental factors extending holding profitability and maximizing excess returns. Notably， the inverse impact of the Z-score is highly pronounced; the significance， directional consistency， and robustness of these three variables are strongly validated through robust cross-comparisons across all regression specifications. This evidence demonstrates that firms with lower Z-scores (higher financial risk)， lower P/B ratios (market undervaluation)， and smaller market capitalizations yield superior investment performance. This study concludes that turnaround stock abnormal returns are jointly driven by risk premia and market undervaluation， and screening via P/B ratio， firm size， and Altman′s Z-score effectively optimizes investment win rates and return performance.