SSACD – Singular Spectral Average Convergence/Divergence

Analog of the MACD indicator based on the “Caterpillar” method of Singular Spectral Analysis (SSA).

Version with limitations of the SSACD Forecast indicator . The restrictions relate to the set of parameters and their range.

SSA is an effective method for processing non-stationary time series with an unknown internal structure. The method allows you to find previously unknown periodicities of a series and make a forecast based on the detected patterns, including trends and oscillations of various scales.

Unlike the MACD, which uses moving averages, the results of processing the CSA do not have a time delay relative to the original series. Accordingly, the indicator is spared from such a typical drawback as lag and more accurately and synchronously reflects the variability of the behavior of the price series. The constructed forecast for the selected “fast” and “slow” price fluctuations takes into account the set of detected multi-scale factors that form the “wave” behavior of a number of data and can be used to reduce risks in the strategy.

The typical behavior, signals, and interpretation of the indicator are similar to the properties of the linear MACD.

[spoiler title=”Read More…”]

Explanation of the set of parameters

The high frequency limit limits the contribution of RF noise to the overall variance of a range of prices. For a fast oscillating average, it is 0.5 – 1.5, for a slow one 1.5-4. The value depends on the length of the fragment. All the oscillations whose contribution does not exceed this level will be filtered out.

BacwardShift is designed to shift calculations along a series of data in order to compare the forecast with known prices and select the indicator parameters.

P.S. Visualization of individual trends along the price chart is available using the SSA Trend Predictor indicator.

[/spoiler]