Date of Award

12-2018

Degree Type

Creative Project

Degree Name

Master of Science (MS)

Department

Economics and Finance

Committee Chair(s)

Tyler Brough

Committee

Tyler Brough

Committee

Todd Griffith

Committee

Ben Blau

Abstract

This paper analyses the performance of the Volatility Index Dynamic Average Indicator (VIDYA) as a method for technical trading. The question was whether or not the buy and sell signals generated by VIDYA could allow a trader to outperform the benchmark rate of return. The strategy is implemented in a similar way to a standard moving average crossover where two lines are charted: a short period VIDYA and a long period VIDYA. The four combinations of VIDYA were used were as follows: 6 with 21 periods, 9 with 21 periods, 12 with 21 periods, and 21 with 50 periods. When the shorter period VIDYA is above the longer period this signals that the underlying asset should be bought. Conversely, when the long period VIDYA is above the short period VIDYA this generates a short signal for the underlying asset. A historical back test was conducted on daily data for SPY and FXE to generate a daily return for trading with the VIDYA. The returns were compared to a benchmark of a buy and hold strategy for SPY and FXE, respectively. A relative performance variable was calculated by taking the difference between the loss of the benchmark and the loss of the strategy. As pointed out in their paper, “Data-Snooping, Technical Trading Rule Performance, and the Bootstrap” by Sullivan et al, data snooping is a serious problem when conducting a back test solely on historical data. I controlled for data snooping via bootstrapping by conducting an out of sample test using Hansen’s SPA test. Rather than bootstrapping the underlying price data and running the back test on that, the bootstrap was conducted on the relative performance variable. These two methodologies should be equivalent but the latter is computationally less intense. None of the trading strategies were found to yield a significant return over the benchmark. At the end, this paper recommends further analysis that should be done.

Share

COinS