Market Impact Model

The irreversible price change caused by the information conveyed by a trade. It reflects the market's belief that the order signals private information about the asset's true value.

• Linear Function: Often modeled as a linear function of the signed trade volume.
• Information Leakage: A large buy order permanently raises the price as the market adjusts to the new information equilibrium.
• Kyle's Lambda: A classic measure of permanent impact, representing the price change per unit of net order flow.

The transient price concession required to attract liquidity and execute a trade quickly. This cost dissipates after the order is completed as the price reverts to its new equilibrium.

• Liquidity Demand: Represents the premium paid for immediacy of execution.
• Resiliency: The speed at which the price reverts after a liquidity shock.
• Decay Function: Typically modeled with an exponential decay or power-law function over time.

A widely observed empirical relationship where market impact costs are proportional to the square root of the trade size.

• Concave Function: Impact increases with size, but at a decreasing rate.
• Universal Scaling: Holds across different asset classes, markets, and time periods.
• Formula: Cost ≈ Spread + σ * (Q / V)^(1/2), where σ is volatility, Q is order size, and V is average daily volume.

A foundational optimal execution model that formalizes the trade-off between market impact and timing risk.

• Trader's Dilemma: Executing quickly minimizes timing risk (volatility exposure) but maximizes market impact. Executing slowly does the opposite.
• Risk Aversion Parameter: A coefficient (λ) that balances the cost of impact against the variance of implementation shortfall.
• Efficient Frontier: Generates an optimal trading trajectory that minimizes a linear combination of expected cost and its variance.

A class of models that view market impact as a propagating perturbation to the order book over time, often used in high-frequency settings.

• Transient Impact Model (TIM): Assumes the impact of each trade decays exponentially, and the total impact is the sum of past impacts.
• Hawkes Processes: Used to model the self-exciting nature of order flow and its decaying impact on mid-price.
• Kernel Estimation: A non-parametric method to estimate the decay kernel directly from tick data without assuming a specific functional form.

A real-time predictor of short-term price movement based on the ratio of resting buy to sell limit orders near the best bid and offer.

• Volume Imbalance: (Bid Volume - Ask Volume) / (Bid Volume + Ask Volume).
• Predictive Signal: A high positive imbalance predicts upward price pressure and higher market impact for a buy order.
• Depth Weighting: More sophisticated models weight volume by its distance from the mid-price, giving more importance to nearby liquidity.

The post-trade quantitative framework used to benchmark execution quality. TCA decomposes actual costs into components like spread capture, market impact, and timing risk. It provides the empirical feedback loop to calibrate and validate market impact models by comparing pre-trade estimates against realized costs, ensuring the model accurately reflects the firm's specific execution footprint.

Strategies that use market impact models as their core objective function to minimize the total cost of trading. These algorithms solve for the optimal trajectory that balances the trade-off between immediate impact (trading aggressively) and timing risk (waiting). Classic solutions like the Almgren-Chriss framework directly parameterize temporary and permanent impact to derive dynamic slicing schedules.

An execution strategy that aggressively hunts for hidden liquidity across dark pools and lit venues to minimize information leakage. Unlike schedule-based algorithms (VWAP/TWAP), these strategies use real-time market impact estimates to decide when to switch from passive resting orders to aggressive liquidity-taking orders, dynamically adapting to avoid signaling large parent order intent.

The fundamental decomposition within any market impact model. Temporary impact reflects the transient liquidity premium paid to absorb limit orders and dissipates quickly post-trade. Permanent impact represents the information content of the trade that permanently shifts the equilibrium price. Distinguishing between these two effects is critical for calibrating execution algorithms and accurately attributing costs.

The risk that a counterparty possesses superior information about an asset's true value. In market impact modeling, adverse selection is the primary driver of permanent impact—the more informed the flow, the larger the permanent price shift. Execution models must account for the probability that aggressive orders are being filled against informed traders who anticipate directional price moves.