EMA Momentum
Reversal Backtest
A controlled historical backtest of the DHA-DEMO-EMA-MOM strategy on RELIANCE.NS, testing execution behavior, costs and strategy performance across 72 daily bars.
What we tested
Experiment 002 tested the EMA Momentum Reversal Alpha strategy using Dhanvi's backtesting engine on 72 daily RELIANCE.NS bars. Orders were modeled using next-bar-open execution and the simulation included commission and slippage.
Experiment configuration
Deterministic configuration parameters supplied to the Dhanvi simulation runtime.
Backtest pipeline
End-to-end execution flow modeling historical inputs through to portfolio settlement.
Backtest result
Verifiable metrics recorded across the 72-bar simulation run.
This is a historical simulation produced by Dhanvi’s backtesting engine. It does not represent live trading performance or expected future returns. Backtests are subject to assumptions and limitations including data quality, execution modeling, commissions, slippage, liquidity and market behavior. Past or simulated performance does not guarantee future results.
Simulated equity transition
Start-to-finish capital reconciliation across the backtested window.
Trade outcomes
Disaggregated breakdown of completed trade executions during the historical test.
Execution costs
Modeled transactional costs applied to order fills.
Direct calculation: ₹78.75 commission + ₹39.37 slippage = ₹118.12 modeled execution costs. These represent simulated frictional modeling, not actual live broker receipts.
What happened
The strategy completed two simulated trades during the tested dataset. Neither trade produced a winning outcome under the configured backtesting assumptions.
Starting from ₹100,000 of simulated capital, the backtest finished at ₹99,116.66, representing a net simulated return of -0.88%.
Maximum observed drawdown during the backtest was -0.91%.
Why publish a losing experiment?
Dhanvi Experiments is intended to document the research process, not only favorable outcomes.
Experiment 002 produced a negative simulated result. Publishing it preserves the complete experimental record and provides a baseline for future strategy iterations.
The next question is not whether this result can be hidden.
The useful question is: What should the next experiment test?
What Experiment 002 does not establish
This experiment does not establish:
- expected future performance
- live trading profitability
- strategy robustness
- performance across different securities
- performance across different market regimes
- statistical significance
Follow the research record.
Experiment 001 tested multi-agent coordination. Experiment 002 tested a single strategy inside our backtest engine. More experiments will be published as Dhanvi develops.