Technical Analysis on Volatility (My Conceptual Framework)
Does technical analysis work? Take any of the countless number of
technical indicators and confirm their effectiveness by simulating the trading
rules over sufficient period of time and one would come to a disappointing
conclusion. Justifiably so, most people
would say technical analysis does not work.
I believe the reason why technical
analysis, in general, has not been so effective is because technical analysis
has focused on analyzing Price and/or Volume over Time. Yet, these arise from the outcomes of the
continuous interaction and intersection of buyers and sellers in a market. By definition, the moment price occurs, it
is already in the past. In other words,
technical analysis and indicators on a chart show us what has happened, not
what will or is likely to happen. In
hindsight, it is always easy to know the outcomes of a decision and go back and
choose what would have been the “right” decision.
Perhaps, it is time to reapply
technical analysis to what it is really good for. Technical analysis, indicators and charts are really good tools
for perception. Markets, price, volume,
time, trend, etc. are all metaphysical objects. Ask the question, “What is a trend?” and you would get countless
answers, which is the same as no one answer.
This is akin to asking the question “What is love?”. In addressing the process of perception, I came up with my
own philosophical concept to more objectively perceive any object, physical or
metaphysical.[1]
The key to greater understanding and
awareness is to perceive anything in multiple or higher dimensions. Once again, technical analysis is a great
tool for perception. If we apply this tool
to volatility (a different, perhaps higher dimension of price) I believe we can
get a much better understanding of price and utilize this understanding for
profit.
Profitability of buying or selling a
stock is a function of volatility. With
zero volatility, zero profits.
Furthermore, when perceiving price through volatility, one does not have
to try to predict the direction of price, but rather if there would be
sufficient price movement in order to realize a profit. In fact, in options/volatility trading, one
could also potentially profit from little or no volatility.
I tested these conceptual assumptions
with a volatility trading strategy on the SPDR S&P500 (SPY) ETF. I utilized two volatility indicators and
applied technical analysis tools on them for timing of when to buy an options
straddle and when to sell an options iron butterfly with outside strikes 5%
away from atthemoney strike.
Additional details include using nearest month options with expiration
at least greater than 10trading days away.
Commissions cost of 0.75 per contract and slippage of 0.05 for the entry
& exit trades were included. The
contract size was set at 10 straddles and 10 iron butterflies. Also, no delta hedging was involved. Finally, the simulation was conducted from
beginning of 2012 until August 2013 expiration, the first entry trade posted
on 2/14/2012. The simulation was
conducted with thinkBack, a backtesting tool on the thinkorswim platform
using end of day marks on options prices.
Strategy

Profits (Gross/Net)
STDEV of Daily Net Profits Change
Potential % Return

Buying next month straddle on
expiration day (cyan)

$10,470 / $8,130
$642
(1) 56.9%

Selling next month iron butterfly
w/outer strikes 5% away on expiration day (yellow)

$18,055
/ $20,935
$463
(2) 100%

Selling iron butterfly w/outer
strikes 5% away using volatility timing (red)

$5,790 / $2,910
$164
(3) 57.0%

Buying straddle, Selling iron
butterfly using volatility timing (blue)

$9,420 / $4795
$510 (4) 57.1% 
(1)
minimum account size required in hindsight would have been
$14,285 to cover for $7,285 maximum loss and $7,000 to initiate a position.
(2)
Surprisingly over the time period, selling an iron butterfly
at the start of each expiration month would have led to consistent losses.
(3)
A minimum of $5,100 would have been required in hindsight.
(4)
A minimum of $8,400 would have been required in hindsight.
The chart and the table shows
simulated results and are for research and education purposes only. More research would be required on longer
time periods, out of sample data, and many more underlying instruments and
options on those instruments in order to test the efficacy of technical
analysis on volatility. Yet, the
strategy to buy and sell options based on volatility indicators seems to show
promise.
The volatility indicator used for
long straddle entry is my Realized Volatility indicator. Details of the Realized Volatility indicator
is discussed in my post “FairVolatility (VIX) Model & Indicator, part I”. The volatility indicator used for short iron butterfly entry is
the Standard Deviation indicator in Metastock software.
Logically, when Realized Volatility is seen
increasing, one would want to buy the straddle in order to take advantage
of potential increase in implied
volatility and expected price movement. On the sell side, one would want to sell an iron butterfly if the directional
movement of the underlying is expected to slow down or be range bound. No distinction was made in this
simulation as to the relative levels of implied volatility vs. realized
volatility but could be parameters for further research.
More specifically, the rules were:
1)
Buy straddle when the criteria for Realized Volatility moving
up was triggered. Simultaneously exit any open short iron butterfly position.
2)
Sell iron butterfly when the criteria for standard deviation
indicator moving down was triggered. Simultaneously
exit any open long straddle position.
3)
Exit any open short iron butterfly position if criteria for
standard deviation indicator moving up was triggered. (Usually, the two indicators moved in same direction with
Realized Volatility leading standard deviation. There were few instances where Realized Volatility remained in a
down move but standard deviation started to move up. This happened in prolonged up move in SPY prices. In these instances, the short iron butterfly position
was exited w/o corresponding long straddle trigger.)
The technical analysis indicators used for timing on these volatility indicators for long and short entry will not be mentioned here, because I am certain through continued research that more effective timing indicators can be found.