How FMZ Simulates Tick Data from Candlesticks
Summary
The document explains FMZ’s simulation-level backtest, where a strategy runs through repeated polling and receives market data through platform API calls. It distinguishes this approach from real market-level testing, which uses recorded tick data, and says tick-based strategies may be represented more realistically with actual ticks.
For simulation, the system derives a sequence of price ticks from each underlying candlestick’s open, high, low, close, and volume. The supplied algorithm uses the candle’s shape and price relationships to choose intermediate prices and timestamps, then places them across the bar. The document also notes that the underlying candle period should generally be shorter than the strategy’s requested candle period, or the generated observations may distort the resulting data.
The method provides a moving price path for testing, but those ticks are inferred from bar data rather than observed market events. The description does not establish that the simulated path captures every possible intrabar sequence, execution condition, or source of live-market slippage.
Key ideas
- FMZ simulation-level backtests repeatedly run strategy logic and return API data as if the program were operating over time.
- Simulation-level testing constructs tick sequences from underlying candlestick values.
- Real market-level testing uses recorded tick observations and can better represent tick-dependent strategies.
- The underlying candle interval should be shorter than the interval requested by the strategy to reduce data distortion.
- Generated ticks are modeled from OHLCV bars and do not reproduce all live market behavior.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.