VIX futures are a type of derivative that allow investors to bet on the future direction of the CBOE Volatility Index (VIX). The VIX is a measure of market uncertainty and is often referred to as the “fear index.” VIX futures allow investors to hedge against or speculate on future changes in the VIX.
Reference [1] presented a systematic approach for trading the VIX futures,
We propose a new approach for trading VIX futures. We assume that the term structure of VIX futures follows a Markov model. Our trading strategy selects a position in VIX futures by maximizing the expected utility for a day-ahead horizon given the current shape and level of the term structure. Computationally, we model the functional dependence between the VIX futures curve, the VIX futures positions, and the expected utility as a deep neural network with five hidden layers. Out-of-sample backtests of the VIX futures trading strategy suggest that this approach gives rise to reasonable portfolio performance, and to positions in which the investor will be either long or short VIX futures contracts depending on the market environment.
An interesting aspect of this paper is that it made use of a utility function to generate trading signals. The authors also performed thorough out-of-sample testing using the k-fold cross-validation technique,
A standard procedure for in-sample training and out-of-sample testing is straightforward: divide the data into two blocks, with the first block designated for in-sample training, and the second block designated for out-of-sample testing. More specifically, we take the VIX futures curves from April 14th of 2008 to August 7th of 2019 for in-sample training, and then utilize the remaining curves from August 8th of 2019 to November 5th of 2020 for out-of-sample testing. But this out- of-sample test is based on a single portfolio run, which means that good performance could be attributable to luck. Therefore, to make full usage of the data, we apply the method of the k-fold cross-validation.
Applying a rigorous out-of-sample testing procedure is necessary in trading system design. This step is, however, often ignored by system developers.
References
[1] M. Avellaneda, T. N. Li, A. Papanicolaou, G. Wang, Trading Signals In VIX Futures, 2021, https://doi.org/10.48550/arXiv.2103.02016
Further questions
What's your question? Ask it in the discussion forum
Have an answer to the questions below? Post it here or in the forum

LONDON — An investor group committed to climate change and controlling $11 trillion in assets has banned members from counting carbon removal schemes towards their emissions reduction targets before 2030, amid increasing scrutiny of the fast-growing market for carbon offsets. The Net Zero Asset Owner…

LONDON — British grocery inflation hit a record 16.7% in the four weeks to Jan. 22, dealing another blow to consumers battling an escalating cost-of-living crisis, industry data showed on Tuesday. Market researcher Kantar said grocery inflation was at its highest since it started tracking…

PARIS — France managed to eke out meager growth in the final quarter of 2022 as falling energy imports and firm business investment offset a pullback in consumer spending, preliminary gross domestic product (GDP) figures released on Tuesday showed. The euro zone’s second-biggest economy grew…

COPENHAGEN, Denmark — Aegir Insights, a sector leader in offshore wind intelligence and investment solutions, is proud to welcome Ørsted A/S (Ørsted), the world’s largest offshore wind developer, as a client. Aegir Insights provides a competitive edge to offshore wind investors by combining deep industry…
