Fill Simulation

Accurately replays the limit order book (LOB) at each price level to determine fill probability. The engine must reconstruct the bid/ask volume available at every tick, not just the top-of-book quote.

• Volume-at-Price: Checks if the aggregated resting orders at a given limit price can absorb the simulated order's size.
• Hidden Liquidity: Models iceberg orders and dark pool reference prices if the historical data feed supports hidden volume flags.
• Tick-Level Granularity: Replays every insertion, cancellation, and execution event to avoid time-interpolation bias.

Simulates the first-in, first-out (FIFO) or pro-rata matching logic of the target exchange. A fill is not guaranteed just because the price trades through your level; your order must reach the front of the queue.

• Time Priority: Assigns a virtual entry timestamp to the simulated order to determine its rank relative to the historical volume traded at that price.
• Partial Fills: Calculates the exact executed quantity when the traded volume is less than the total queue size.
• Exchange-Specific Rules: Models the distinct matching algorithms of venues like CME (FIFO) vs. Eurex (pro-rata).

Differentiates between liquidity-taking and liquidity-providing order fills. A marketable order sweeps the book, while a passive limit order joins the queue.

• Market Orders: Consume available volume across multiple price levels, calculating the volume-weighted average price (VWAP) of the sweep.
• Limit Orders: Only fill if the market price trades through the limit (for buys, trades at or below the limit price) and queue position is exhausted.
• Crossed Markets: Handles the edge case where a simulated limit order is placed inside an already crossed national best bid and offer (NBBO).

Injects realistic execution latency between the signal generation timestamp and the order's arrival in the market. This delay causes slippage, the difference between the decision price and the actual fill price.

• Deterministic Delay: Applies a fixed or stochastic delay (e.g., 500 microseconds) to simulate network and processing latency.
• Adverse Selection: Models the probability that a quote fades or the book thins out during the latency window.
• Tick-by-Tick Drift: Captures the price movement during the delay using the actual historical quote sequence.

Simulates the logic of a Smart Order Router (SOR) that splits a parent order across fragmented exchanges and dark pools to minimize total execution cost.

• Venue Sweeping: Routes the order to the venue with the best available price first, then to the next, based on a configurable fee schedule.
• Reg NMS Compliance: Models the protected quote status of US exchanges, ensuring the simulation does not trade through a better displayed price.
• Fill Probability Decay: Assigns a lower fill probability to dark pool and non-displayed venue slices based on historical hit rates.

Outputs granular fill data to feed a Transaction Cost Analysis (TCA) module. Every simulated fill must carry its cost attribution for post-trade evaluation.

• Explicit Costs: Tags each fill with the exact commission, exchange fee, and rebate based on the venue and order type.
• Implicit Costs: Calculates the implementation shortfall by comparing the fill price to the arrival price at signal time.
• Market Impact Decomposition: Separates the fill price drift into temporary (liquidity) and permanent (information) impact components.