本研究以台灣加權指數期貨為研究對象，探討均線趨勢交易策略於不同風險控制機制下之績效表現與資本效率差異。研究首先比較多組移動平均線參數組合，包括MA(5，10，20) 、MA(5，10，60) 、MA(10，20，60) 與 MA(20，60，200) ，藉以篩選較具代表性之趨勢交易模型。實證結果顯示，MA(10，20，60) 在累積報酬、盈虧比與最大回撤之間具備較佳平衡，因此作為後續風險控制機制比較之基礎策略。 在風險控制方面，本研究分別導入 Buffer Ratio 固定式價格濾網與Average True Range （ATR）動態停損機制，比較不同參數設定對策略績效之影響。研究結果發現，Buffer 機制能有效降低均線糾結所造成之頻繁交易問題，其中Buffer Ratio = 0.02 具有較佳之累積報酬表現；而ATR 動態停損則能依市場波動程度自動調整停損距離，其中ATR 倍數為4 時，在累積報酬率、回撤控制及資產修復能力方面展現較佳績效。 此外，本研究進一步探討不同槓桿倍數下之策略表現，分析固定式與動態式風控機制於槓桿環境中的資本效率與風險承受能力。結果顯示，適當之風險控制機制不僅能提升策略長期穩定性，亦能改善槓桿交易下之資產增長品質與回撤控制能力。 綜合而言，本研究證實風險控制機制對均線趨勢交易策略之長期績效具有顯著影響。研究結果顯示，Buffer 機制在提升累積報酬方面具有較佳效果，而ATR 動態停損則在回撤控制、資產修復能力及市場適應性方面表現較佳。兩種風險控制方式各具優勢，投資人可依報酬追求與風險承受能力選擇適合之策略架構。;This study investigates the performance and capital efficiency of moving average trend-following strategies under different risk control mechanisms using Taiwan Stock Index Futures as the research target. First， several moving average combinations， including MA(5，10，20)、MA(5，10，60)、MA(10，20，60) and MA(20，60，200)， are examined to identify the most representative trend-following model. The empirical results indicate that MA(10，20，60) achieves the most balanced performance in terms of cumulative return， profit-loss ratio， and maximum drawdown， and is therefore selected as the benchmark strategy for subsequent analyses. To evaluate the effectiveness of risk management mechanisms， this study incorporates both the Buffer Ratio， a fixed-price filtering approach， and the Average True Range (ATR)， a dynamic volatility-based stop-loss mechanism. The results show that the Buffer Ratio effectively reduces excessive trading caused by market noise and moving average whipsaws. Among the tested settings， a Buffer Ratio of 0.02 generates superior cumulative returns. In contrast， the ATR stop-loss mechanism dynamically adjusts the stop-loss distance according to market volatility， demonstrating greater adaptability to changing market conditions. The findings reveal that an ATR multiplier of 4 provides the best overall performance in terms of cumulative returns， drawdown control， and recovery capability. Furthermore， this study examines strategy performance under different leverage levels to assess capital efficiency and risk tolerance. The results suggest that appropriate risk control mechanisms can significantly improve the long-term stability of trend-following strategies while enhancing capital growth and drawdown management in leveraged trading environments. Overall， the findings confirm that risk management plays a critical role in the long-term performance of moving average trading strategies. Compared with the fixed Buffer mechanism， the ATR-based dynamic stop-loss approach demonstrates superior market adaptability and risk control capability， providing valuable implications for the design and practical implementation of trend-following trading systems. Keywords: Moving Average Strategy， Trend Following， Buffer Ratio， Average True Range (ATR)， Risk Management， Leveraged Trading， Capital Efficiency
