signal-processing

Analyze financial time series with Fourier and wavelet methods for signal extraction.

10|2|Updated Mar 14, 2026
One-click install
npx skills add https://github.com/brainbytes-dev/everything-claude-trading --skill signal-processing
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: signal-processing
Source: https://github.com/brainbytes-dev/everything-claude-trading/tree/main/skills/quant-methods/signal-processing
Command: npx skills add https://github.com/brainbytes-dev/everything-claude-trading --skill signal-processing

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Financial data is often contaminated with noise, making it hard to identify reliable signals. This skill provides a structured set of methods to denoise, decompose, and interpret financial time series, enabling robust indicators from price data.

Core Features & Use Cases

  • Fourier transforms for cycle detection and spectral analysis to identify dominant frequencies in financial data.
  • Wavelet and EMD based multi-scale decomposition to extract multi-timeframe signals (trend, swing cycles, and noise) with denoising options.
  • Practical denoising and filtering techniques (HP filter, moving averages, Savitzky–Golay) with guidance on lag and phase considerations.
  • Use Case: Decompose a daily price series to isolate the long-run trend, mid-term cycles, and high-frequency noise for strategy signals.

Quick Start

Instruct the agent to perform a multi-scale signal analysis on a price series to extract trend and cycles.

Frequently Asked Questions about signal-processing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I denoise financial time series data to extract trading signals?

Financial time series denoising applies filtering techniques like HP filters, Savitzky-Golay, and wavelet decomposition to separate underlying trends from market noise, yielding robust indicators. The skill provides guidance on lag and phase considerations to ensure signal reliability.

What is multi-scale decomposition for price series and when do I need it?

Multi-scale decomposition uses wavelet and EMD methods to split a price series into long-run trends, mid-term swing cycles, and high-frequency noise. You need it when single-timeframe indicators fail to capture complex market dynamics across different horizons.

Can I use Fourier transforms for cycle detection in intraday data?

Fourier transforms support cycle detection and spectral analysis on both intraday and daily financial data to identify dominant frequencies. The skill includes validation checks and parameter guidance to ensure accurate frequency extraction for live trading contexts.

Does this signal processing approach support non-leaky causal processing for live trading?

Yes, the signal processing methods enforce non-leaky causal processing designed specifically for live trading environments. This ensures that denoising and decomposition operations only use historical data points, preventing future information from bleeding into current indicator calculations.

What's the best way to isolate long-term trends from high-frequency noise in macroeconomic series?

The best approach combines HP filtering and wavelet-based multi-scale decomposition to isolate long-term trends from high-frequency noise in macroeconomic series. These methods offer denoising options while maintaining phase integrity for accurate cycle detection.

Why does my moving average indicator lag and how can signal processing help?

Moving averages inherently introduce lag and phase shifts in indicator signals. The skill addresses this by offering alternative denoising techniques like Savitzky-Golay filters and wavelet methods, providing guidance on balancing noise reduction against lag for timely trading signals.