Two practical related papers to #198 - Exploiting Term Structure of VIX Futures Monday, 15 June, 2015

#198 - Exploiting Term Structure of VIX Futures

Authors: Donninger

Title: Selling Volatility Insurance: The Sidre- and Most-Strategy

Link: http://www.godotfinance.com/pdf/VIXFuturesTrading_Rev1.pdf

Abstract:
This working-paper examines and improves a VIX-Futures calendar-spread strategy proposed in the literature. The strategy relies on the typical term-structure of VIX-futures. Additionally a naked short-selling strategy is considered. The strategies have similar  characteristics to selling Puts on the S&P-500. There is some risk, but also a lot of fun.  The strategies are  an interesting alternative investment-vehicle to boost the performance of a fund.

and ...

Authors: Donninger

Title: VIX Futures Basis Trading: The Calvados-Strategy 2.0

Link: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2379985

Abstract:

I developed in a previous working paper the Sidre and Most-Strategy. The strategy relies on the typical termstructure of VIX futures. The Calvados is a refined and condensed version of these strategies. The starting point was a paper of Simon and Campasano. The authors demonstrate that the VIX futures basis does not have significant forecast power for the change in the VIX spot index, but does have forecast power for subsequent VIX futures returns. It is especially profitable to short VIX futures contracts when the basis is in contango. The original Calvados working paper presented improved metrics and parameter settings of the Simon&Campasano approach. The current working paper improves the original work in several points and extends the historic backtest. The overall patterns of the original results are reassured and improved upon. The Calvados is traded in the Sybil-Fund. It is so far the pick of the bunch. One gets a lot of fun for a medium dose of risk.

Notable quotations from the academic research papers:

...

"The VIX can not simply be replicated. Additionally the VIX is mean-reverting. The VIX volatility is almost an order higher than the S&P and explodes in a market-crash. Hence VIX-futures (and options) behave very different to commodity- or stock-index futures. VIX-futures trade in normal times at a premium to the VIX. There is also a typical term- structure. The premium increases with maturity like the positive part of a logistic function.

R. Rhoads proposes in chap. 12 of [Rhoads, R.: Trading VIX Derivates, John Wiley&Sons, 2011] a simple calendar-spread strategy which is based on the normal term-structure. Sell with a maturity of 3-4 months the nearer future and go the next future long. The position is closed on Friday (5 days) before expiry. The strategy   sells volatility insurance. It is similar to selling Puts. The second future hedges somewhat the risk. The strategy is – like all insurance-selling strategies – in the long run profitable. Rhoads proposes a simple filter. One stays out of the market as long as the future- premium is negative. The filter improves risk (but not necessarily the overall win) somewhat.

The Sidre-1 Strategy is an implementation of this idea. One enters with a maturity of N-days a calendar spread. The trade is only done if the near-term short future trades at a premium to the VIX. The relation of the second future is not considered directly. One waits as long as the condition is met. But the maturity must be at least 10-days. One defines additionally a danger-zone. Per default the danger zone is as long as the initial-maturity. But it can be shorter. A danger-zone of 0 corresponds to the strategy proposed by Rhoads in [Rhoads, R.: Trading VIX Derivates, John Wiley&Sons, 2011]. If the futures fall within the danger-zone below the VIX, the position is closed. The relation between the futures-price and the VIX is used as a stop-loss-trigger."

"In [D. Simon, J. Campasano: The VIX Futures Basis: Evidence and Trading Strategies. June 27, 2012] the authors consider only the most nearby future. But according to the analysis, it is sometimes advantageous to trade futures with longer maturity. These futures are less sensitive to sudden spikes of the VIX. With a similar daily roll one would certainly prefer the longer maturity. The tail risk is much lower.

The trading-strategy in [D. Simon, J. Campasano: The VIX Futures Basis: Evidence and Trading Strategies. June 27, 2012]  is: One selects the most nearby future if the daily roll is above a given enter-threshold. The future is held a maximum of 5 trading days. It is sold back premature, if the daily-roll falls below the stop-threshold or if there are less than 10 trading-days till maturity. The Calvados strategy modifies these rules. One selects from all the futures the best one (with the largest daily-roll). The maximum maturity is 93 trading days (about 4 months). For longer maturities volume is low and hence the bid-ask-spread increases. The best daily-roll must be as in the original paper larger than the enter-roll. But the position is hold as long as the daily-roll is above the stop-threshold. If the maturity is less than 10 trading days, the future is sold back."


Are you looking for more strategies to read about? Check http://quantpedia.com/Screener

Do you want to see performance of trading systems we described? Check http://quantpedia.com/Chart/Performance

Do you want to know more about us? Check http://quantpedia.com/Home/About