資產基礎類型因子分析法下的可轉債套利型對沖基金套利策略獲利性實證

Abstract

資產基礎類型因子分析法（Asset-Based Style Factor Analysis）不但可以用來衡量對沖基金的風險與報酬特性，同時，資產基礎類型因子分析法下的資產類型因子（Asset-Based Style Factor）本身也是一種可以實際應用的投資策略。 本研究主要在介紹資產基礎類型因子分析法如何用以衡量對沖基金的風險與報酬特性，並且以國內可轉債市場的資料對資產基礎類型因子分析法下的可轉債套利策略：正資金成本差額策略（Positive Carry）、信用風險套利策略（Credit Arbitrage），以及波動性風險套利策略（Volatility Arbitrage）等消極策略的獲利性及可轉債市場的獲利來源進行實證。 實證結果發現，以國內的可轉債市場而言，正資金成本差額策略與波動性策略的風險調整性報酬要比同時期的股價指數或所有可轉債構成的投資組合都來得好。正資金成本差額策略與波動性策略的累積報酬也鮮少為負，呈現出對沖基金策略追求絕對報酬的特性。信用風險套利策略則因為研究期間一連串的財務危機事件使得信用價差擴大而績效不佳。然而，在策略的實際操作上，由於國內公司債市場不夠活絡，因此波動性套利策略在對信用風險從事避險時必須要用到較客製化的產品。這對套利者而言，會產生極大的困擾。因此發達公司債市場是國內發展所有固定收益相關產業的首要之務。

Not only does the asset-based style factor analysis provide a solution to the measurement of the risk and return characteristics of hedge funds, but the style factors employed by asset-based style factor analysis are practically investable strategies. To the risk and return characteristics of hedge funds, we introduce how asset-based style factor analysis can serve to measure the intrinsic risk and return characteristics of a particular category of hedge funds. To the strategies employed by asset-based style factor analysis, we pick up the passive strategies dedicated to the convertible arbitrage hedge, which are the positive carry strategy, credit arbitrage strategy, and volatility arbitrage strategy, to exam the profitability of these investable strategies in the convertible bond market of Taiwan. The empirical result indicates that the risk-adjusted return of the positive carry strategy and volatility arbitrage strategy are superior to those of the stock composite index and convertible bond market portfolio comprised of all trading convertible bonds. The cumulative returns of the positive carry strategy and volatility arbitrage strategy are rarely negative which is in line with the investment philosophy of hedge funds to pursue absolute return. In response to a series of financial distressed episodes, the performance of credit arbitrage strategies are deemed by the performance of the aforementioned benchmarks.