EVT Crash Gauge: Advanced Tail-Risk Measurement

Related: crash spike trade pattern - another powerful indicator for MetaTrader 5 traders.

Overview

Many standard MetaTrader 5 risk tools assume that the maximum potential loss is similar to the largest loss observed historically. Tools like Average True Range adjust position sizes based on recent volatility, Monte Carlo simulations reorder past trades, and Historical Value at Risk uses percentiles from your return history. These methods focus on the main body of the distribution and assume future market movements will not exceed past extremes.

Also recommended: projection bands - similar indicator with strong performance on MetaTrader 5.

However, financial returns exhibit fat tails, meaning extreme events occur more frequently and are larger than predicted by a normal distribution. Relying solely on past losses will systematically underestimate the potential impact of rare crashes.

The normal distribution model (curve) compared to fat-tailed financial returns (histogram). The shaded left tail represents the uncaptured risk.

This indicator introduces Extreme Value Theory (EVT), a branch of statistics designed to model these tail events, directly into MQL5. EVT focuses specifically on the tail of the distribution, using a theoretical shape justified by limit theorems rather than assumptions. This fitted model can then extrapolate potential losses far beyond historical data, estimating events like a 99.9% loss or a once-in-three-years market move, even if such an event has never occurred in your recorded history.

The Method: Peaks-Over-Threshold

The indicator employs the Peaks-Over-Threshold (POT) method. It converts chart closing prices into a series of one-sided downside losses, recording the magnitude of each loss and setting gains to zero. A high threshold (u) is selected based on a quantile of this loss series. Losses below this threshold are discarded, and only the *exceedances* (the amounts by which losses surpass the threshold) are analyzed. The Pickands-Balkema-de Haan theorem states that for many underlying distributions, these excesses converge towards a Generalized Pareto Distribution (GPD) as the threshold increases. This means the shape of the tail does not need to be guessed; it can be fitted to the data.

Peaks-Over-Threshold illustration: a threshold is set on the loss series, exceedances above it are captured, and a Generalized Pareto Distribution is fitted to these excesses.

The fitting process uses maximum-likelihood optimization of the GPD's shape and scale parameters. This is performed using the ALGLIB constrained optimizer included with the terminal, eliminating the need for external libraries, Python bridges, or DLLs.

Scope of Application

EVT is a tool for measuring risk *magnitude*, not for market *timing*. It quantizes the potential severity of rare losses and indicates whether the tail risk is increasing or decreasing. It does not predict the timing of crashes or suggest trading directions. Interpreting a rising tail shape as a

You may also like: vgrid line custom - excellent alternative for indicator users on MetaTrader 5.

#property copyright "RobotFX"
#property link "https://robotfx.org"
#property version   "1.00"
#property strict
#property indicator_separate_window
#property indicator_buffers 1
#property indicator_plots   1
#property indicator_label1  "EVT VaR"
#property indicator_type1   DRAW_LINE
#property indicator_color1  clrOrangeRed

.......

• This tool is provided for educational and testing purposes only.
• Past performance does not guarantee future results.
• Trading involves substantial risk of loss. Use on a demo account first.
• Results may vary depending on market conditions, broker, and settings.
• We recommend thorough backtesting and forward testing before using with real funds.