MarketCycles

This package provides digital signal processing indicators developed by John F. Ehlers.

Currently the original indicators shared in his book: Cycle Analytics for Traders, Advanced Technical Trading Concepts are provided with intent to explore the DSP space and provide new intuitions based on the Ehlers framework.

Outside contributions are welcome!

Available Indicators

• Available Indicators

  • Supersmoother
  • Decycler
  • Decycler Oscillator
  • Band Pass Filter
  • Hurst Coefficient
  • HP-LP Roofing Filter
  • Zero Mean Roofing Filter
  • Roofing Filter
  • Modified Stochastic
  • Modified RSI
  • Autocorrelation (Multiple Lag Matrix)
  • Autocorrelation (Single Lag)
  • Autocorrelation Periodogram
  • Autocorrelation Reversals

• TO DO

  • Dominant Cycle
  • DFT Spectral Estimate
  • Comb Filter Spectral Estimate
  • Adaptive RSI
  • Adaptive Stochastic Indicator
  • Adaptive CCI
  • Adaptive Band Pass Filter
  • Even Better SineWave Indicator
  • Compute and Display Convolution
  • Classic Hilbert Transformer
  • Hilbert Transformer Indicator
  • Dominant Cycle Using the Dual Differentiator Method
  • Dominant Cycle Using the Phase Accumulation Method
  • Dominant Cycle Using the Homodyne Method
  • Fisher Transform to the Adaptive RSI Indicator
  • SwamiCharts RSI
  • SwamiCharts Stochastic

Usage

Pkg.clone("https://github.com/flare9x/MarketCycles.jl")

Each indicator function requires an input of a single dimension array of Float64 type. Each indicator function has its own set of arguments which change the level of some parameter. This may be the length of a look back period, the high or low pass periods or bandwidth value. For example we may call @doc function_name to see the associated documentation for the specific indicator function:

julia> @doc AutoCorrelationReversals
Autocorrelation Reversals - Equation 8-3

  The indicated reversals are very sensitive to the smoothing of the price data.
 ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡

  LPLength is made available as an indicator input to decrease or increase the number of
indicated reversals as desired.
 ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡

  The AvgLength parameter is also made available as an indicator because this averaging also
impacts the number of indicated reversals.
 ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡

  Care should be taken when increasing the value of this input because the lag of the
indicator increases in direct proportion to the increase of the value of the AvgLength.
 ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡

  Typical delay of the indicator will be about three bars when the AvgLength parameter is set
to a value of 3.
 ≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡≡

AutoCorrelationReversals(x::Array{Float64}; min_lag::Int64=1, max_lag::Int64=48,
LPLength::Int64=10, HPLength::Int64=48, AvgLength::Int64=3)::Array{Float64}

Call the function as below for lags 1 to 48:

AutoCorrelationReversals(your_data,min_lag=1,max_lag=48,LPLength=10,HPLength=48,AvgLength=3)

We may see what this looks like over dummy data:

using MarketCycles
using Gadfly

# Generate dummy data
srand(121)
n = 1000
op = 100.0 + cumsum(randn(n))
hi = op + rand(n)
lo = op - rand(n)
cl = 100.0 + cumsum(randn(n))
index = collect(1:1:length(cl))
for i = 1:n
 if cl[i] > hi[i]
 	cl[i] = hi[i]
 elseif cl[i] < lo[i]
 	cl[i] = lo[i]
 end
end

# Apply autocorrelation reversals function
auto_cor_reversals = AutoCorrelationReversals(cl,min_lag=1,max_lag=48,LPLength=10,HPLength=48,AvgLength=3)

# Plot
white_panel = Theme(
 panel_fill="white",
 default_color="blue",
 background_color="white"
)
p1 = plot(x=index,y=cl,Geom.line,
Guide.xlabel(nothing), Guide.ylabel("Price"), Guide.title("Dummy Data"),white_panel)
p2 = plot(x=index,y=auto_cor_reversals,Geom.line,Guide.xlabel("Time Index"),Guide.title("Autocorrelation Reversals"), Guide.ylabel("Autocorrelation Reversals"),white_panel)
out = vstack(p1,p2)

# Save Plot
draw(PNG("C:/Users/Andrew.Bannerman/Desktop/Julia/auto_correlation_reversals.png", 1500px, 800px), out)

For the output:

One may line up the reversals to the turning points within the data!

Available Function Names

SuperSmoother, Decycler, Decycle_OSC, BandPassFilter, DominantCycle, HurstCoefficient, HPLPRoofingFilter,
    ZeroMeanRoofingFilterP0, ZeroMeanRoofingFilterP1, RoofingFilterIndicator,
    ModifiedStochastic, ModifiedRSI, AutoCorrelationIndicator, SingleLagAutoCorrelationIndicator, 
    AutoCorrelationPeriodogram, AutoCorrelationReversals

Feel free to explore any of the functions with:

@doc SuperSmoother