We test the Vix as a filter in a Mean Reverting strategy on the’s & P 500 – Technical analysis and trading system

To build an effective trading strategy, it is important to identify favorable market conditions and define clear rules to operate. In this study we will use a particular filter: the Vix, also known as the index of fear.

The goal is to develop a trading strategy on the Future of the S&P 500 (ES), evaluating whether the integration of the VIX can improve its performance.

Do you want help to improve your trading approach? Click here >>>

What is the Vix and because it is called “Fear index”

The Vix (Volatility Index) is an indicator that measures the implicit volatility of the options on the’s & p 500. It is often defined as an index of fear because it tends to rise in moments of strong uncertainty in financial markets.

As shown in Figure 1, the VIX has an inverse correlation with the stock market: when the S&P 500 descends, the Vix tends to climb, and vice versa. This happens because in periods of turbulence volatility increases, while in moments of stability the market becomes more predictable and the VIX goes down.

Given this behavior, the VIX can be a useful tool for filtering trading operations. But does it really add value to a systematic strategy? We will find out in the next paragraphs.

Development of a Mean Reverting strategy on the Future of’s & P 500

We will proceed by developing a Mean Reverting type Long strategy. This means that we start from the hypothesis that if a price suffers an excessive movement compared to its usual trend, the market will tend to reabsorb it over time.

For this reason, we will open a Long position in a hyper -fired phase, when the price breaks the minimum of the previous session, and we will close the position in a hyper -computer condition, or when the price breaks the maximum of the previous session.

Subsequently, we will add a simple trend filter, i.e. a mobile average of the closing prices calculated on the last 50 bars. We will only work when the price is above the mobile average, in order to avoid operations in strongly bearish market phases.

Finally, to control the risk and eliminate any outliers (anomalous values), we will implement a 2,500 dollar stop and a 5,000 dollar take profit. This will allow us to manage operations in a more balanced way, avoiding that extreme movements influence the results of the strategy.

Performance analysis: Net Profit, DrawDown and strategy metrics

Let’s now analyze the performance of the strategy from 2010 to date.

The Equity Line in Figure 2 shows that the strategy has produced a positive result in the long run, with a net profit of about 90,000 dollars. However, we notice how stability has decreased in recent years, with more marked drawdowns and greater volatility in the results. This suggests that although the approach has worked well in the past, recently it may have encountered greater difficulties in maintaining the same effectiveness.

Observing the total trade analysis in Figure 3, we note that the strategy performed 712 operations, with a successful percentage of 70%. This means that in most cases the strategy has closed the trade in profit.

Another relevant figure is the Averal Trade, equal to about 127 dollars. This value is already large enough considering the slippage and commissions, but it is definitely improved.

Use of Vix as a volatility filter to improve strategy performance

We can use the VIX to filter the entrances of the strategy, introducing a further level of selection that takes into account market volatility. There are several ways to do it: for example, we could only operate if the VIX is above or under a certain predefined level, or if it is greater or lesser than its mobile media.

However, for this article we will use a very simple and effective approach. Specifically, we will analyze whether the daily closure of the VIX is greater or lesser than its opening. In other words, if during the session the VIX has increased, it means that volatility is expanding, while if it has fallen, it indicates a contraction of volatility.

The filter will be applied in this way:

• If the closing of the Vix is ​​less than its opening, i.e. volatility has decreased, we will proceed with a purchase order to the next bar.
• If, on the other hand, the closure of the Vix is ​​higher than its opening, the trade will not be performed.

This method allows us to operate only when volatility is reducing, avoiding entrances to contexts of uncertainty or panic selling.

Strategy results operating in case of contraction of volatility

After applying the filter based on the daily variation of the VIX, the results of the strategy have worsened significantly.

The Equity Line in Figure 4 shows a clear decrease in the Net Profit, which has drastically reduced to the previous version of the strategy. In addition, the drawdown is much more violent, with deeper and prolonged loss phases over time. This indicates that, instead of improving the stability of the strategy, the filter had the opposite effect, excessively penalizing the operations.

Turning to the total trade analysis in Figure 5, we can clearly see the consequences of the introduction of the filter:

• The number of operations dropped to 449 trade, a sign that many revenues have been filtered.
• The success percentage fell to about 67%, a worsening compared to the version without filter.
• The Averal Trade dropped drastically to about 24 dollars, a value too low to make the strategy sustainable over time, especially considering slips and commissions.

In other words, the filter destroyed the strategy.

At this point, to check if volatility can still offer a useful contribution, we will try to apply the opposite condition: operate when the closure of the VIX is greater than its opening, or on the days when volatility has increased.

Strategy results by operating in case of expansion of volatility

After applying the opposite condition, then on days of increasing volatility, the results of the strategy have changed drastically positively.

The Equity Line in Figure 6 shows a clear improvement compared to both previous versions. The Net Profit exceeded $ 100,000, and the curve appears more stable, with reduced Drawdown than the original strategy. In addition, growth is more constant, without the long stages of difficulties seen in the last period of the backtest.

Analyzing the Total Trade Analysis in Figure 7, let’s see that:

• The success percentage increased to about 73%, a better value than the version without filter and, without doubt, compared to the approach based on the reduction of volatility.
• The Averrage Trade is more than doubled, reaching about 243 dollars.

These results confirm that, contrary to what one might think, the strategy works better on days of increasing volatility rather than in those of reduction. A possible reason is that the strategy is already limited to operating only when the price is above the mobile average, excluding many strongly bearish market scenarios. In this way, there is a tendency to intercept growing volatility phases but with a bearish trend not yet too advanced, conditions in which the rebound is more likely.

Considerations on the effectiveness of the Vix as a volatility filter in Mean Reverting strategies

In the common imagination, we tend to think that operating in low volatility phases is the best choice, especially for Mean Reverting strategies. This logic certainly has its foundation: more peaceful markets can offer more controlled and less exposed entrances to extreme movements.

However, with this study we have seen that this is not always the case. In the case of a Mean Reverting strategy, operating when there was a price shock, like a panic selling, can represent an advantage. After a violent movement, in fact, the probability of witnessing a rebound is evidently higher.

At this point, the strategy could be further refined, which however, could already be ready for live trading.

Do you want help to improve your trading approach? Click here >>>

Until next time,

Andrea Unger