# Options Strategy – Volatility

### Video Tutorial

**In this section of beeTrader the Chain Options acquired are often used as illustrated in the section Volatility Surface . Therefore, it is advisable to carry out the acquisition before proceeding with the analyzes below.**

### Volatility Index

Introduction: the volatility indices represent the implicit volatility mediated by a series of options on different expiries. There are indices in which the implied volatility they represent is the rolling one, that is the volatility of series of options which also imply different expiries but which do not exceed 30 days.

For example, on the first day of the year the index will only represent the implication of January, while on January 15th the implication we will read will be the average of the options that have 15 days of life on January and 15 days of life on February.

As you can see it will always be 30 days.

Other types of calculations always represent implicit volatility of options but grouped in different ways.

** The beeTrader Volatility Index has a proprietary calculation mode that differs from how the other indices are calculated and therefore absolutely not COMPARABLE measures **.

We devised this algorithm because there was no Volatility Index that indicated not so much the implied volatility that I already find in the other indices, but gave an indication (Index) of the probable trajectory of the underlying. We have extended it to all the instruments dealt with and you can find it on stocks, currencies, futures and bonds.

** Use **: beeTrader’s Volatility Index is used both to evaluate the trend and to identify the period of rising or falling volatility. There are two display modes which are:

Difference

This method allows the assessment of the difference between the Volatility Index and the historical volatility of the security with reference to 30,50,75,100,150 periods, or the difference of each individual historical volatility with another of a different period.

The analysis with the Volatility Index is important to be able to define if the Market Maker is pricing risk (implied volatility) greater or lesser than the historical volatility and therefore we know if we will be dealing with discount or increased options.

The analysis of historical volatility is instead important to understand the trend and the level of value of historical volatility.

In the example the difference between the Volatility Index (selected in the Reference to Compare menu) is displayed with the historical volatilities at 30 days and 50 days displayed in the upper graph.

Oscillator

The Volatility Index can also be displayed as an oscillator, this type of display makes it easier to read and consequently decide whether you are in high or low volatility.

Consider the empty spaces left by the indicator as shown and in this way you will avoid many false signals.

Given that you are creating a volatility-trend relationship, you need to understand that it will be long-term and not for daily trading. It takes up position and is maintained until the opposite signal.

An example of using the volatility index oscillator is the one shown in the image.

### Options Smile

Options Smile shows the volatility smiles of the call and put options of the surface stored for the underlying, in this regard the date to which the displayed smileys refer is indicated. If no curve is present it is necessary to acquire a surface using the Volatility Surface tool accessible from the menu.

### Skewness

Skewness allows real-time differences in the slope of the skewness to be evaluated on the desired deadlines. Shows by default the volatility skew of the stored surface for the underlying. If no curve is present the skew are flat and therefore it is necessary to acquire a surface using the Volatility Surface tool accessible from the menu .

It is composed of three graphs, in each of which the deadline to be seen must be set using the appropriate command on the menu. In each of the three graphs (therefore in each of the three chosen deadlines) it is possible to analyze the volatility obtained from the memorized surface, or activate the realtime (from the top menu) in order to compare the volatility of the options stored with the current one.

The area below the graphs is dedicated to the comparison between stored and current skewness. This study is based on the term structure with which the options on currencies in particular are listed, but also on indices and stocks. This structure assumes a slope (skewness) which is always negative as the market tends to price the risk more in a Bear trend than a Bull.

In fact from the images we see that the strikes (Put OTM and Call ITM that would be those affected Bear trend) below the ATM value have higher volatilities than those above.

With this configuration, always negative of the slope, we can assume that greater slope is equal to greater fear of bear trend and less slope instead reflects a bullish expectation. It is therefore an excellent forecast tool since the slope anticipates the movement of the underlying as it is determined by the market makers who are well aware of the short-term trend.

### Cone & Probability

- The table corresponding to the number 1 serves to know how much the historical movement of the underlying was when a certain number of days were missing. So if I read on the X24 axis and then I go to the 3% box I read 70, this means that the underlying when it was 24 days to expiry has made a movement of 3% in 70% of the time. With such information I can think of constructing a figure whose gain derives from the exceeding of the strikes. In detail, on the left edge of the table there are the days relating to the available deadlines (whose range can only be included from 2 days to a maximum of 365). In our case, today, we have deadlines of 24, 59, 87, 150, etc. days). In the top row of the table we find pre-set values of% price deviation (1.5 / 2 / 2.5 / 3 / etc), on the basis of which the calculations we find in the colored boxes are made. In the boxes in red the internal number expresses the times in which the displacement was less than 50%, in yellow the times in which the displacement is between 50% and 75% and in green the times in which the displacement was greater than 75%.
- Selecting a row on graph 1, in graph 2 (the historical graph of the title) horizontal lines appear. The red lines identify a possible strangle sold, the green lines a possible strangle bought, the distances from the ATM of the strikes that make up the strangles are indicated under the historical graph.
- Graph 3 displays the volatility cone: it is a tool with which the historical volatility of the underlying is compared with the implicit one for the various strikes and expiries. The cone is built using only the expiries up to one year, the historical volatility of all the points between 2 and 365 periods is used and then the average is drawn for a year drawn with the blue line. Once the average is obtained, the standard deviations are always calculated at one year, in orange 1 stdev and in green 2 stdev. Implied volatilities for each maturity are represented with dots for calls and with triangles for put. The result is being able to verify the implied volatilities that are inside the cone that have normal values, while those outside the cone have high or low implied volatility values depending on whether they are above or below the cone.

### Vega IN / OUT

Vega IN / OUT was designed and developed to take advantage of the retracements from the volatility peaks identified by Volatility Index.

** In the literature we read that volatility is sold when it is high and bought when it is low. But this information causes heavy losses. In reality, volatility is sold when it folds down and bought when it is going up. It may seem like a small difference, but it is substantial. This highlights the Vega IN / OUT tool! **

An example of use is to exploit the volatility peaks to sell a Straddle when the Volatility Index crosses down the 2 line, so when it is returning from a very high volatility. In this case we try to sell Call and Put on a deadline (1.5 – 2 years) with a low delta, around 0.2. In doing so the figure will be little influenced by the movements of the underlying and will draw its gain from the lowering of the vega.

In the event that the underlying moves a lot and therefore unbalances the delta, thus negating the gain given by the vega, it may make sense to intervene on the delta by unbalancing the figure, then selling another option (Call or Put depending on the direction had of the underlying). In this way the figure will be unbalanced because there will be more contracts sold on one side, but the delta will be balanced.

A level to which balancing can be done is to wait for the ratio between the loss of one side / the gain of the other side to be = 2.