Skip to content
MeanyDeanyQuantitative Research Systems

MSc Economics thesis · University of Copenhagen

Volatility Regime Filtering in Futures Markets

An intraday NQ and ES futures study using EGARCH as a volatility-regime risk and admissibility layer—not as a direction predictor.

View thesis PDF
Markets
NQ · ES
Frequency
Daily model · 5-minute evaluation
Model role
Risk and admissibility
Context
Academic research

Research scope

Volatility context without directional overreach

The thesis asks whether an econometric volatility layer can improve the discipline of an otherwise comparable intraday research framework.

Daily EGARCH estimates are aligned to five-minute futures observations. The regime classification determines whether the volatility environment is considered admissible for the academic framework; separate intraday logic carries the directional hypothesis.

This separation makes the model's contribution testable. The same underlying intraday logic can be compared with and without EGARCH conditioning instead of crediting every result to the volatility model.

Research tags

Academic foundation
Financial econometricsEGARCHNQ futuresES futuresIntraday dataRisk filtering

Research question

Can an EGARCH filter add value without predicting direction?

Model input
Daily futures returns
Research output
Conditional-volatility regime
Intraday context
Five-minute NQ and ES data
Comparison
Filtered versus no-filter framework

Methodology

Three layers with separate responsibilities

  1. Volatility layer01
    • EGARCH(1,1) with Student-t innovations
    • Daily conditional-volatility estimates
    • Percentile-based volatility-regime classification
    • Asymmetric response to signed return shocks
  2. Intraday layer02
    • Five-minute intraday NQ and ES observations
    • Daily regime labels aligned to intraday dates
    • Trend-following and mean-reversion research modules
    • Transaction-cost and risk-control assumptions
  3. Validation layer03
    • EGARCH-filtered versus no-filter ablation
    • Alternative volatility filters
    • Walk-forward evaluation
    • Bootstrap, subperiod, and market-level robustness

Framework architecture

A traceable path from data to comparison

  1. 01

    Data

    Daily and five-minute NQ and ES futures observations

  2. 02

    Volatility

    EGARCH conditional volatility and regime labels

  3. 03

    Admissibility

    Volatility context governs whether exposure is considered

  4. 04

    Direction

    Separate intraday logic handles directional hypotheses

  5. 05

    Validation

    Ablation, robustness, and walk-forward comparisons

Results interpretation

What the evidence supports—and what it does not

Supported interpretation
  • The thesis reports that EGARCH conditioning contributed useful risk and admissibility context in the main academic setting.
  • Ablation helps isolate the filter's contribution from the underlying intraday logic.
  • Volatility models can be more useful as conditioning layers than as directional predictors.
Unsupported interpretation
  • EGARCH does not establish price direction.
  • The evidence is not causal proof or a universal futures rule.
  • Academic backtests do not provide live-trading permission or investment advice.

Robustness

Limitations stay attached to the finding

Sample dependence
Findings remain specific to the studied samples, instruments, and research design.
Threshold choice
Volatility-regime thresholds are design choices that require sensitivity checks.
Cost assumptions
Transaction costs and risk controls materially shape intraday evaluation.
Walk-forward degradation
Re-selection under realistic chronology can weaken apparent in-sample performance.
Market differences
NQ and ES should be interpreted individually as well as together.
Academic boundary
The framework is a thesis experiment, not an execution system.

Capabilities demonstrated

  • 01Financial econometrics
  • 02EGARCH volatility modeling
  • 03Intraday futures data
  • 04Ablation and robustness
  • 05Walk-forward validation

Research boundary

Academic evidence, not execution permission

This page describes an academic research project. It is not a live trading system, does not provide signals or execution, and is not investment advice.