由於金融科技日漸成熟，智能投資及程式交易也越來越發達，再加上近幾年全民編程的火熱，越來越多投資者運用程式對金融商品進行分析及操作，因此券商也趕著這波商機開發下單API供投資者使用，但將回測交易策略實際投入市場操作仍有許多安全機制需考量，甚至會有滑價問題之產生，因此本論文會實際運用券商下單API建立自動化交易系統，以及實證三種不同的下單策略對於交易策略績效的影響，試圖找到績效偏離原始回測策略最小之最適下單方法。
經實證結果發現，隔日低於開盤價之最低賣價買，五日後收盤前5分鐘最高買價賣的表現最佳，每筆交易策略大約都會產生0.2%~0.5%報酬率之滑價成本；於收盤前五分鐘之最高買價賣的績效沒有較收盤之最高買價賣好；加入最低賣價比開盤價低才買的限制績效略較佳但會導致無法成交。

Due to the maturity of Fintech, intelligent investment, program trading and a lot of people learning program, more and more people begin to analyze financial products with program. Brokage also find the opportunity to develop the API for investors to place order with program. However, there are still many safety mechanisms to be considered when using back-testing trading strategy in market and there may even be slippage problems. Therefore, this papers will actually use the brokerage order API to establish automation trading system and evidence the performances of three different order placement strategies to the trading strategies. Try to find out the optimal order placement method with the smallest performance deviation from the original backtesting strategy.
The empirical results show the best order placement strategy that buying at the lowest ask price lower than the opening price the next day, and selling at the highest bid price 5 minutes before the close five days later; The performance of selling at the highest bid price five minutes before the close is not better than selling at the highest bid price at the close; Adding the restriction of buying only when the lowest ask price is lower than the opening price is slightly better, but it will result in failure to trade.

論文審定書 i
摘　　要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 vii
程式碼目錄 viii
第一章、緒論 1
1.1、 研究動機 1
1.2、 研究目的 1
1.3、 研究架構與流程 2
第二章、文獻探討 4
2.1、券商下單API發展 4
2.2、相關文獻 4
2.2.1、訂單價格群聚效應(Order Price Clustering) 4
2.2.2、訂單不平衡(Order Imbalance) 5
2.2.3、訂單相關研究 5
第三章、研究方法 6
3.1、券商下單API功能簡介 6
3.1.1、使用資格及使用限制 6
3.1.2、市場資料(Market Data) 7
3.1.3、下單(Order) 10
3.1.4、帳戶資料(Account Data) 11
3.2、下單前預備 11
3.3、下單程序及偵測訂單狀態 13
3.3.1、登入及匯入憑證 15
3.3.2、開啟委託回報及成交回報輸入Mysql執行續 15
3.3.3、訂閱標的五檔報價及成交資訊 15
3.3.4、偵測每檔標的有五檔報價及成交資訊 15
3.3.5、下單判斷 16
3.3.6、判斷是否下單買進賣出 16
3.3.7、執行下單及偵測訂單狀態 16
3.3.8、執行續偵測 18
3.3.9、交易結果紀錄及彙整 18
3.4、模擬交易之建構 18
3.4.1、模擬訂單 18
3.4.2、模擬委託回報 19
3.4.3、模擬成交回報 19
3.5、交易策略之回測 20
3.5.1、蒐集回測數據 20
3.5.2、交易策略來源 20
3.5.3、下單策略之建構 21
第四章、實證與結果 22
4.1、下單策略實證結果 22
4.1.1、交易策略一 22
4.1.2、交易策略二 24
4.1.3、交易策略三 25
4.1.4、交易策略四 27
4.2、小結 29
第五章、結論與建議 30
5.1、結論 30
5.2、建議 30
參考文獻 31
附錄 32

Lien, D., Hung, P. H., Zhu, J. D., & Chen, Y. H. (2019). Price limit changes and market quality in the Taiwan Stock Exchange. Pacific-Basin Finance Journal, 55, 239-258.
Aitken, M., & Frino, A. (1996). The determinants of market bid ask spreads on the australian stock exchange: Cross‐sectional analysis. Accounting & Finance, 36(1), 51-63.
Johnson IV, B. H., & Watson, E. D. (2018). The Explanatory power of order imbalance measures. Quarterly Journal of Finance, 8(01), 1850003.
Chu, X., & Qiu, J. (2021). Forecasting stock returns using first half an hour order imbalance. International Journal of Finance & Economics, 26(3), 3236-3245.
Handa, P., & Schwartz, R. A. (1996). Limit order trading. The Journal of Finance, 51(5), 1835-1861.
Ahn, H. J., Bae, K. H., & Chan, K. (2001). Limit orders, depth, and volatility: Evidence from the stock exchange of Hong Kong. The Journal of finance, 56(2), 767-788.