Multicollinearity of Technical Indicators

Dr. Rajeev Shukla

www.lifefunetc.com
[email protected]
Investrade group, Indore
• All my best thoughts were stolen by the
ancients.
― Ralph Waldo E erso

• Originality is nothing but judicious imitation.

The most original writers borrowed one from
another.
― Voltaire
• Pure price systems are close enough to the
North Pole that any departure tends to bring
you farther south.
- William Eckhardt
mul·ti·col·lin·ear·i·ty
noun \"+kəˌlinēˈarə̇tē, -kä-\
???
The existence of such a high degree of
correlation between supposedly
independent variables being used to
estimate a dependent variable that the
contribution of each independent variable
to variation in the dependent variable
cannot be determined.
John Bollinger

"A cardinal rule for the successful use of

technical analysis requires avoiding
multicollinearity amid indicators.
Multicollinearity is simply the multiple
counting of the same information. The use of
four different indicators all derived from the
same series of closing prices to confirm each
other is a perfect example."
Problems with multicollinearity

• In some sense, the collinear variables contain

the same information about the dependent
variable. If nominally "different" measures
actually quantify the same phenomenon then
they are redundant.
• A principal danger of such data redundancy is
that of over-fitting in regression analysis
models.
What indicators to choose?

1000+ collection of Technical indicators for

trading needs to be simplified and made clear.
Types of indicators (Source: Metastock)

• Trend indicators
• Volatility indicators
• Momentum indicators
• Cycle indicators
• Market strength indicators
• Support and resistance indicators

(Few indicators may fall in more than one category.)

 Trend Indicators (Source: Metastock)
Indicator Creator Indicator Creator
Aroon Chande Price Oscillator
Commodity Selection Index Wilder Qstick Indicator Chande
DEMA Mulloy r-squared
Directional Movement Wilder Raff Regression Channel Raff
Forecast Oscillator Chande Standard Deviation Channel Equis
Linear Regression Indicator Standard Error
Linear Regression Slope Standard Error Bands Equis
Linear Regression Trendline (Andersen)
MACD Appel Standard Error Channel Equis
Moving Averages (all TEMA Mulloy
calculation methods) Time Series Forecast
Parabolic SAR Wilder Trendlines
Performance Vertical Horizontal Filter White
Polarized Fractal Efficiency Hannula Zig Zag Merril
Volatility Indicators (Source: Metastock)
Indicator Creator Reference
Average True Range Wilder Average True Range
Bollinger Bands Bollinger Bollinger Bands
Commodity Selection Index Wilder Commodity Selection Index
Moving Average (variable) Chande Variable
ODDS™ Probability Cones Fishback ODDS Probability Cones
Relative Volatility Index Dorsey Relative Volatility Index
Standard Deviation Standard Deviation
Standard Error Bands Equis (Andersen) Standard Error Bands
Volatility, Chaikin's Chaikin Volatility, Chaikin's
Volatility, Option Bookstaber Option Volatility
 Momentum Indicators (Source: Metastock)
Indicator Creator Reference Indicator Creator Reference
Accumulation Swing Accumulation Swing Random Walk Index Poulus Random Walk Index
Wilder
Index Index
Chande Momentum Chande Momentum Range Indicator Weinberg Range Indicator
Chande
Oscillator Oscillator Relative Momentum Relative Momentum
Commodity Channel Lamber Commodity Channel Altman
Index Index
Index t Index Relative Strength
Dynamic Momentum Dynamic Momentum Relative Strength Index Wilder
Chande Index
Index Index Stochastic Momentum Stochastic
Intraday Momentum Intraday Momentum Blau
Chande Index Momentum Index
Index Index
Stochastic Oscillator Lane Stochastic Oscillator
Linear Regression
Linear Regression Slope
Slope Swing Index Wilder Swing Index
MACD Appel MACD TRIX unknown TRIX
Mass Index Dorsey Mass Index Ultimate Oscillator Williams Ultimate Oscillator
Momentum Indicator Momentum Williams' %R Williams Williams' %R
Price Oscillator Price Oscillator Williams'
Williams' Accumulation/
Williams Accumulation/Distribu
Price Rate-Of-Change Appel Price Rate-Of-Change Distribution
tion
Cycle Indicators (Source: Metastock)
Indicator Creator Reference
Cycle Lines Cycle Lines
Detrended Price Oscillator Detrended Price Oscillator
Fibonacci Time Zones Time Zones
Fourier Transform Fourier Transform
MESA Sine Wave Indicator Ehlers MESA Sine Wave
 Market Strength Indicators (Source: Metastock)
Indicator Creator Reference Indicator Creator Reference
Accumulation/ Accumulation/ Negative Volume Negative Volume
Williams Fosback
Distribution Distribution Index Index
On Balance
Demand Index Sibbet Demand Index On Balance Volume Granville
Volume
Chaikin Money
Chaikin Money Flow Chaikin Open Interest Open Interest
Flow
Chaikin A/D Positive Volume Positive Volume
Chaikin Oscillator Chaikin Fosback
Oscillator Index Index
Price Volume
Ease of Movement Arms Ease of Movement Price Volume Trend
Trend
Herrick Payoff Equis Trade Volume
Herrick Payoff Index Herrick Trade Volume Index
Index (Slauson) Index
Klinger Oscillator Klinger Klinger Oscillator Volume Volume
Quong/Soud
Money Flow Index Money Flow Index Volume Oscillator Volume Oscillator
ack
Moving Average Volume Rate-Of- Volume Rate-Of-
Moving Average Change Change
Arms Calculation
(volume adjusted)
Methods
Support & Resistance Indicators (Source: Metastock)

Indicator Creator Reference

Andrews' Pitchfork Andrews Andrews' Pitchfork
Envelope Envelope
Fibonacci Arcs, Fans,
Fibonacci Studies
Retracements
Gann Lines, Fans, Grids Gann Gann Studies
Projection Bands Widner Projection Bands
Projection Oscillator Widner Projection Oscillator
Quadrant Lines Quadrant Lines
Speed Resistance Lines Speed Resistance Lines
Tirone Levels Tirone Tirone Levels
Trendlines Trendlines
Advantages of unrelated
indicators
• The best regression models are those in which the
predictor variables each correlate highly with the
dependent (outcome) variable but correlate at most
only minimally with each other.
• Such a model is often called "low noise" and will be
statistically robust (that is, it will predict reliably
across numerous samples of variable sets drawn
from the same statistical population).
Use complimentary indicators

• To avoid the problems associated with

multicollinearity, traders should select
indicators that work well with, or
complement, each other without providing
redundant results.
Few combinations suggested

• Standard deviation
+
• MACD
+
• OBV
Few combinations suggested

• MA crossover
+
• MFI
Few combinations suggested

• Bollinger bands
+
• RSI
+
• Volume Oscillator
Few combinations suggested

• PSAR
+
• ROC
+
• Trendlines
Few combinations suggested

• PSAR
+
• MFI
+
• Trendlines
A word about Market Breadth
Indicators
• Arms Index (TRIN)
• Advance-Decline Line
• Advance-Decline Volume Line
• Bullish Percent Index
• High-Low Index
• McClellan Oscillator
• McClellan Summation Index
• Net New Highs
• Percent Above Moving Average
• Record High Percent
• Volatility Index (VIX)
Eccentric indicators (Bizarre!)
Just for fun!!! (Source:CNBC)
• Hemline/Skirt Length Indicator
• Super Bowl Indicator
• Billboard Top 100 Indicator
• Sports Illustrated Swimsuit Cover
• The Big Mac Index
• Lipstick Indicator/Lipstick Effect
Take home message
• Use optimization and backtesting but avoid
overoptimization as past never equals future.
• Technical analysis deals in probabilities rather than
certainties. There is no combination of indicators
that will accurately predict the markets' moves 100%
of the time.
• While too many indicators, or the incorrect use of
indicators, can blur a trader's view of the markets,
traders who use technical indicators carefully and
effectively can more accurately pinpoint high-
probability trading set-ups, increasing their odds of
success in the markets.
Suggested Reading

• Technical Analysis of the Financial Markets: A

Comprehensive Guide to Trading Methods and
Applications (New York Institute of Finance) -
John J. Murphy
• Technical Analysis of Stock Trends - John
Magee
• Technical Analysis: The Complete Resource for
Financial Market Technicians -Charles D.
Kirkpatrick II, Julie Dahlquist
A special thanks -

• Investrade group, Indore and my dear friends

Nareshji, Deveshji and Himanshuji for their
help and cooperation.
• ATMA for allowing me to speak on this
platform.
Every battle is won
before it is ever fought.
—Sun Tzu
THANK YOU