GARCH Volume Volatility [MarkitTick]

Title: GARCH Volume Volatility

Description

Overview
The GARCH Volume Volatility (GV) indicator is a sophisticated quantitative tool designed to analyze the rate of change in market participation. While the vast majority of technical indicators focus on Price Volatility (how much price moves), this script focuses on Volume Volatility (how unstable the participation is).

Market volume is rarely distributed evenly; it tends to cluster. Periods of high activity are often followed by more high activity, and periods of calm tend to persist. This behavior is known as "heteroskedasticity." This script utilizes an Exponentially Weighted Moving Average (EWMA) model—a core component of Generalized Autoregressive Conditional Heteroskedasticity (GARCH) frameworks—to model these changing variance regimes.

By isolating volume volatility from raw volume data, this tool helps traders distinguish between sustainable liquidity flows and erratic, unsustainable volume shocks that often precede market reversals or breakouts.

Methodology and Calculations

1. Logarithmic vs. Percentage Returns
The foundation of this indicator is the calculation of "Volume Returns"—the period-over-period change in volume.
- The script defaults to Logarithmic Returns. In financial statistics, log returns are preferred because they normalize data that can vary wildly in magnitude (such as cryptocurrency volume spikes), providing a more symmetric view of changes.
- Users can opt for standard percentage changes if they prefer a linear approach.

2. Variance Proxy (Squared Returns)
To measure volatility, the direction of the volume change (up or down) matters less than the magnitude. The script squares the returns to create a "Variance Proxy." This ensures that a massive drop in volume is treated with the same statistical weight as a massive spike in volume—both represent a significant change in the volatility of participation.

3. GARCH-Style Smoothing (EWMA)
Standard Moving Averages (SMA) treat all data points in the lookback period equally. However, volatility is dynamic. This script uses an EWMA model with a tunable "Lambda" (Decay Factor).
- The Recursive Formula: The current calculation relies on a weighted average of the current variance and the previous period's smoothed variance.
- Memory Effect: This allows the indicator to "remember" recent volatility shocks while gradually letting their influence fade. This mimics the GARCH process of conditional variance.

4. Dynamic Statistical Thresholds
The final output is the Volatility (square root of variance). To make this data actionable, the script calculates a dynamic upper and lower limit based on the standard deviation (Z-Score) of the volatility itself over a user-defined lookback period.

How to Use

The indicator plots a histogram that categorizes the market into four distinct volatility regimes:

1. High Volatility (Red Histogram)
Trigger: Volatility > High Band (Upper Standard Deviation).
Interpretation: This signals an extreme anomaly in volume stability. This is not just "high volume," but "erratic volume behavior." This often occurs at:
- Capitulation bottoms (panic selling).
- Euphoric tops (blow-off tops).
- Major news events or earnings releases.

2. Elevated Volatility (Maroon Histogram)
Trigger: Volatility > Mean Average.
Interpretation: The market is in an active state. Participation is changing rapidly, but within statistically normal bounds. This is common during healthy, trending moves where new participants are entering the market steadily.

3. Normal/Low Volatility (Green Histogram)
Trigger: Volatility is within the lower bands.
Interpretation: The market volume is stable. There are no sudden shocks in participation. This is typical of consolidation phases or "creeping" trends where the price drifts without significant volume conviction.

4. Extremely Low Volatility (Bright Green/Transparent)
Trigger: Volatility < Low Band.
Interpretation: The "calm before the storm." When volume volatility collapses to near-zero, it implies that the market has reached a state of equilibrium or disinterest. Historically, volatility is cyclical; periods of extreme compression often lead to violent expansion.

Settings and Configuration

Core Settings
- Use EWMA: When checked (Default), uses the recursive GARCH-style calculation. If unchecked, it reverts to a simple SMA of variance, which is less sensitive to recent shocks but more stable.
- Log Returns: Uses natural log for calculations. Highly recommended for assets with exponential growth or large volume ranges.
- Length: The baseline period for the calculation.
- Threshold Lookback: The number of bars used to calculate the Mean and Standard Deviation bands.
- EWMA Lambda: The decay factor (0.0 to 1.0). A value of 0.94 is standard for risk metrics.
-- Higher Lambda (e.g., 0.98): The indicator reacts slower and is smoother (long memory).
-- Lower Lambda (e.g., 0.80): The indicator reacts very fast to new data (short memory).

Visuals
- Show Thresholds: Toggles the visibility of the statistical bands on the chart.
- High Band (StdDev): The multiplier for the upper warning zone. Default is 1.5 deviations. Increasing this to 2.0 or 3.0 will filter for only the most extreme events.

Disclaimer This tool is for educational and technical analysis purposes only. Breakouts can fail (fake-outs), and past geometric patterns do not guarantee future price action. Always manage risk and use this tool in conjunction with other forms of analysis.