The Optimal Leverage of Taiwan Index Futures under Kelly Criterion

• Main Authors: Yu-je Ye, 葉育哲
• Other Authors: Chou-Wen Wang
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/69576016235306388309

碩士 === 國立高雄第一科技大學 === 風險管理與保險研究所 === 103 === Which asset to choose and when to invest are complicated issues of trading, because there are many financial instruments in the market. To solve not only complicated but also basic problem, we would like to study a method for simplifying relational problem, which does or does not exist. In this paper, we use Kelly criterion was proposed by John Larry Kelly, Jr. to calculate leverage as trading position. If the asset has high liquidity, we depend on historical return and historical volatility to calculate trading position at each period. The purpose is to chase geometric growth rate of asset, equal to maximize wealth in the end. We use GARCH (1, 1) to forecast volatility and return (Normal, StudentT) then observe the performance of using historical return and historical volatility. We choose Taiwan stock weighted index futures (TX), Taiwan electronic weighted index futures (TE), and Taiwan financial weighted index futures (TF) as our benchmark. To compare that benchmark’s performance base on different in-sample days with intraday data and daily data. In result, GARCH-t model is the best performance of models with 990 in-sample days. The TX’s investment grows to 389.36% of the initial in eight years. Finally, we suppose to fix leverage to 5 or -5 at each period and find benchmark making a large profit when accuracy rate of forecasting leverage is to 56%, TX’s minimum 5 percentile of terminal value is to 643.6%.