Blog

Gold Gone Wild…

TradingPub Admin | September 11, 2012

Responsive image

Special thanks to TradingPub Commentator, Nick Pritzakis, on the following post on Gold Trading Strategies:

Ok, the first chart looks at the realized/historical volatility in Gold futures for the last 20, 40, 60, 80 and 120 trading days. You’ll notice that two estimators are used to measure volatility.

Why? By using two different estimators we are able to gather more information on the actual volatility. For example, the estimator labeled (C) captures volatility using yestarday’s close and today’s close (aka close to close). In addition, the second estimater, which is labeled (YZ), captures the intraday volatility by using the open, high, low and closing prices.

So how can this be useful?

Let’s assume that GC futuers closed 1740 yesterday and closed at 1740 today. The first estimator would show us that realized volatiltiy was 0%

Now, let’s assume that during the trading day GC opened at 1745, had a high of 1765, a low of 1735 and closed at 1745. Based on the example, intraday volatility is greater than overnight risk. Not only that, but the close to close estimator is doing a poor job of capturing true volatility.

This is extremely important information, especially when you have to deal with positions that require adjusting.

Now, the difference between overnight risk (c)  and intraday volatility (YZ) is not overly signficant. For the last 20 and 40 trading days intraday volatility has been greater. This means the market has been wilder during the trading session vs. yesterday’s close to today’s close. In options terms, making adjustments before the close has made more sense than trying to make them during the trading session. However, the relationship changes for the last 60 to 120 trading days.

Usually, we’ll see more randomness in near term volatiltiy. This means we have the potential to see abnormally high or low volatility in the short term…but as more observations are gathered, volatiltiy tends to smooth itself out and those abnormal moves are averaged in.

Think about it this way…if Derek Jeter goes 0 for 5 today, do you think the coaches are going to ask him to change his swing? Do you think the media will start writing about how is batting average is .000 over the last game?

Probably not…as you can see, there is a lot of randomness in the near term. However, some traders will get excited about the volatiltiy of XYZ over the last 10 trading days as if it is a substantial statistic.

The less observations there are…the greater the randomness and sampling error you’ll get from the data.

In order to put the present numbers into perspective, we’ll have to take a look at a volatility cone later.

The next chart looks at the term structure in GC options. When observing implied volatility, the most important options to focus on are the At-The-Money options (ATM)

The ATM implied volatility in gold options increased last week… as gold futures prices moved higher. Unlike equity indices, commodities like gold tend to see a rise in implied volatiltiy when the futures price rises. Of course, this varies amongst products…but it is not unusual in the commodity options market.

Think about it this way…if the fear in equity indices is for prices to decline, than the demand for put options should be greater. However, higher commodity prices means you have less money for other goods and services…hence, the fear is for prices to increase. This would increase the demand for call options.

Of course, the futures options market is a lot more complicated than this basic example…but these dynamics are also what make them more interesting to trade compared to equity options.

The shape of the term strucuture has a positive skew. Now, this is what we’d expect under normal market conditions. The more time we have until expiration, the greater the uncertainty…and hence the higher implied volatility should be.

Next we want to compare the realized volatility to the ATM implied volatility.

Now, the spread between realized and implied volatility is slightly positive. For example, when you compare the last 20 and 40 trading days (realized vol) to the NOV contract the spread is slightly positive…it’s slightly positive when you compare the last 40 and 60 trading days to the DEC contract…it’s slightly positive when you compare the last 60 and 90 trading days to the JAN contract…

You should always take into account the bid/ask spread as well as fees and commissions. Once those are factored in, the spread becomes extremely narrow across the entire curve.

Ideally, you’d like to short volatiltiy when the spread is positive (options are expensive) and go long volatility when the spread between realized and implied is either flat or negative (options are cheap).

However, in order to put the current realized volatity numbers into perspective….it’s best we look at a realized volatility cone.

Realized volatility for the last 20, 40 and 60 trading days are at or below the 25th percent threshold. It’s at or around the historical median for the last 80 and 120 trading days.

Ideally, you’d like go long volatility when the spread between realized and implied is negative or flat…and when realized is relatively low.

Next we’ll be taking a look the fourth moment, aka modern kurtosis.

The Kurtosis cones are used to analyze the flatness or peakedness of a returns distribution…compared to the normal distribution.

According to the sample, the distribution usually has higher peaks around the mean when compared to the normal distribution. In addition, there is a higher than normal probability to see extreme price moves (on both sides, up or down)…AKA fat tails.

The higher kurtosis (positive)…we should expect to see OTM and ITM options have a higher implied volatility than the ATM options. If the kurtosis is negative or flat, we should expect the OTM and deep ITM options to have a similar or lower implied volatility than the ATM options.

We are witnessing significantly positive kurtosis for the last 60, 90 and 120 observations.

Under these conditions, we’d expect to see OTM and deep ITM options have a greater implied volatility then the ATM options.

Which OTM options? The calls or the puts? In order to figure that out we’ll have to look at a skewness chart.

The Skewness cones are used to analyze the symmetry of the distribution.

According to the sample’s median, the distribution is moderately skewed to the left (negative skew). This simply means that we have a longer tail on the left… than what we would expect to see with a normal distribution. That means that the equidistant puts should have a higher implied volatility than the equidistant puts.

Currently, the distribution is moderately skewed to the right for the last 30, 60 and 90 observations. This means that most of the returns are concentrated on the left of the mean…with more extreme moves occurring towards the upside.

Presently, the skew is positive, meaning that call options should be more expensive than put options.

Now, the skew is significantly positive for the last 30 observations and moderately positive for the last 60 and 90 observations…well above the 75th percent threshold, across the entire curve.

Under the present structure, the equidistant OTM calls should have a greater implied volatility then the equidistant OTM puts for the OCT, NOV, DEC and JAN contracts.

The actual distribution vs. the normal distribution over the last 252 trading days.

The skewness and kurtosis cones…along with the distribution chart are used to get a rough estimate of what the implied volatility skew might look like. Unlike volatility, skewness and kurtosis are random and don’t experience the same mean reverting properties…

The implied volatility skew charts compare implied volatility to moneyness.

Now, moneyness is simply the strike price divided by the futures price (adjusted for time).

The charts below provide great information on how the options market moved last week, not only for the ATM, but the ITM and OTM options. It’s important to see how the changes occur amongst the different contract months, strike prices, puts and calls…especially for those who trade calendar spreads.

Gold Strategy Analysis

One could argue that whatever the FED has to say this week has already been factored into the price of gold futures and options. With that said, realized volatility remains relatively low for the last 20, 40 and 60 trading days.

Ok, so the spread between realized and implied is positive, but it’s very narrow…especially when you factor in what you lose from the bid/ask spread and commissions.

Clearly, there is no edge in selling volatility this week.

Long volatility strategies still remain favorable…and options are still on the cheap side.

Remember, near term options are more of a realized volatility/gamma play…long term options are more of an implied volatility/ (vega vs. theta) play.

Now, for those that only like to sell options…your best position is cash.

Directional trades are high risk/high reward this week…the upcoming news announcements could push the market big in one direction (up or down, your guess is as good as mine). With that said, long volatility strategies that start off delta neutral make the most sense.

Strategies to Avoid: naked short calls and puts. Short straddles and short strangles, short ratio spreads

Favorable Strategies:  Long straddle, long strangle, call back spread or put back spread (if you have a directional bias), short condor and short butterfly (short the wings). Reverse Calendar Spread (long more near term options/ short less long term options)

Each week, Nick does a full report for multiple markets including the Bonds, Crude OIl, Gold, Euro and Emini SP 500.  These sheets help determine if options are relatively "cheap" or expensive" and help to give you a cheat sheet outlining Favorable Strategies and the ones to avoid.  If you are interested in learning more, he is currently offering 15% off for TradingPub members at the following link:  Learn More Here.

Cheers,

The TradingPub
“Trade, Talk, Learn – Cheers to Success”
Check out our Breaking News Tab on facebook

Risk Disclaimer: Past performance is not indicative of future results. Futures trading involves substantial financial risk. Views of guest commentators do not represent those of TradingPub.com. Article intended for educational purposes only and not meant in anyway as a solicitation to buy or sell certain securities. Please consult your personal financial adviser before using this information for your own trading purposes.