Discrete Wavelet Transform (DWT)

Category: trend, cycle

An implementation of the Stationary (À trous) Haar Discrete Wavelet Transform. It decomposes price series into a smoothed approximation (trend) and multiple levels of detail coefficients (noise/cycles).

Formula

c_j = (c_{j-1} + c_{j-1}[2^{j-1}]) / 2 \\ d_j = c_{j-1} - c_j \\ \text{where } c_0 = \text{source}, c_j \text{ is approximation at level } j, d_j \text{ is detail at level } j.

Inputs

• Source (default: close)
• Decomposition Levels (default: 4)
• Output Component (default: 0)

See signal primitives and every published strategy that uses Discrete Wavelet Transform (DWT) on WOBR StrategyVerse.