依據中華民國證券櫃檯買賣中心公佈之新金融商品業務概況，我們可以看出債券選擇權雖然到2004年7月才核准開放，但是從年度成交金額，債券選擇權均居所有債券與利率衍生性商品之冠。由於債券選擇權市場仍處於開放初期，相關資料如債券選擇權隱含波動率取得不易，因此關於台灣債券選擇權市場並未如指數選擇權市場吸引很多學術與實務界之注目。 本研究嘗試對於台灣債券選擇權市場，從評價模型、市場實務概況描述及未來可能之發展，作進一步之研究；此外，本文亦參酌Goodman and Ho（1997）所進行之美國債券選擇權損益之實證研究，來探討賣出債券「買進選擇權」之一方，是否能同樣在台灣債券選擇權市場獲得相對等之報酬。 本實證分析方法基本上是假設，「債券買權」賣方賣出履約殖利率為一個基本點價外之「債券買權」，選擇權存續期間是以兩個星期為一個循環，以中華民國證券櫃檯買賣中心公佈之每日10年期指標公債加權平均殖利率作為相對應部位調整之依據，分別代入實際波動率與期初「債券買權」之隱含波動率，進行Delta Neutral避險，且假設債券拆借與資金融通利率為中華民國證券櫃檯買賣中心公佈之等殖成交行情表（附條件）中所揭露之附條件利率，來探討賣方是否能夠賺取一定合理之報酬率及其可能面對之風險。 從2005年4月至2005年11月底之實證期間，共計13個循環週期，我們發現下列幾個現象： ◆債券買權與賣權之隱含波動率普遍高於實際波動率，賣權隱含波動率又高於買權。 ◆以賣出債券買權為例，在不考慮交易成本下，賣出買權之一方均能獲利，但損益之變異性都相當大。 ◆在避險波動率參數之使用上，代入實際波動率，不論是從損益之絕對數值或每單位風險報酬，均優於代入期初買權隱含波動率。 ◆從等殖成交行情表（附條件）中所揭露之附條件利率，即債券融資與融券利率，其波動性遠大於債券殖利率，因此在進行債券選擇權交易時必需將此列入評價之重要考量因素。 ◆如果考慮權利金收入，由於目前稅法無法與避險損失互抵，必需被課徵25%營利事業所得稅，因此選擇權賣方均無法獲利。 / According to the fact reports of OTC derivatives released by GraTai Securities Market (GTSM) in 2005, trading value of bond option was top of all interest rate derivatives and bond derivatives, though it was opened lately in July 2004. Due to the difficulty to get the market information like implied volatility of options, we haven’t seen considerable studies on Taiwan bond option market. Owing to the writer of this paper serves in this field, we can take advantage of collecting these data from Interdealer brokers and dealers. Accordingly, we have ground to develop this research. In this paper, we begin our discussion by presenting the landscape of Taiwan bond option market and then turn to review the basic models. Lastly, we reference the empirical study of US OTC Treasury option done by Goodman and Ho (1997) to construct the framework for measuring how investors are fairly compensated by selling call in Taiwan bond option market. To evaluate the profit and risk of option writing, we assume writer can sell two-week call options on ten-year Treasury note at market implied volatility, with a strike yield at one basis point out of daily closing yield. Option writer uses daily closing yields to do the delta neutral hedge. Besides, we take overnight call loan rates of Electronic Bond Trading System in GTSM as the borrowing and lending rates of government bond. Over the empirical period of 2005.04.1-2005.11.30, several circumstances can be found from this study. 1.In average, the implied volatilities are higher than actual volatility. This is in agreement with the results of Goodman and Ho (1997). Besides, implied volatility of put option is generally higher than call option. 2.Without considering transaction cost, call option writer can have vulnerable profit from selling volatility. 3.Adoption of actual volatility can bring better absolute profit and risk-adjusted return than implied volatility in the option valuation. 4.In this article, it shows a striking effect of the borrowing and lending rates of government bonds on the movement of implied volatility spread between call and put. 5.Under current Act of Income Tax, the hedging loss of option writer cannot be offset by premium income. Therefore, call writer can not make profit after tax.