Frameworks
Last updated
Last updated
The RWG employs a combination of quantitative and qualitative measures to fulfill its many responsibilities—ranging from parameterizing FiRM markets and performing due diligence on new lending opportunities, to assessing liquidity strategies and managing an increasingly complex array of DOLA Fed facilities. Recognizing the intricate, often non-linear interactions between different market parameters (e.g., collateral factors, liquidation thresholds, and borrow limits), we have developed in-house frameworks to guide our decision-making. These models, outlined below, enable us to fine-tune market parameters for optimal performance, while also supporting our broader risk management duties. Their evolution and use cases have been discussed on the as part of our ongoing “Behind the Scenes” series.
The Collateral Parameterization Model maps the interactions between various market parameters, including Supply Ceiling, Collateral Factor, Liquidation Factor, Liquidation Incentive, and Fee. By using simulation data derived from price impacts of the underlying asset, this tool allows us to simulate and analyze the interplay between these parameters. It provides insights into their combined effects on the ecosystem's health, enabling us to set parameters that balance risk and opportunity effectively.
Our Liquidation Factor Model determines the optimal liquidation factor by simulating the total gas spent by a liquidator using platforms like Tenderly. It takes into account the liquidation incentive and other variables to set the most efficient liquidation factor. This model ensures that the liquidation process is cost-effective for liquidators, encouraging their participation and thereby maintaining market stability.
The Daily Borrow Limits Framework leverages comprehensive data extraction and analysis from the largest liquidity pools of the underlying collateral. By analyzing liquidity pool data, we can set daily borrow limits that prevent excessive borrowing and mitigate risks associated with liquidity crises or market manipulation. This proactive approach helps in maintaining a balanced borrowing environment that safeguards both the protocol and its users.
The Risk Observer Checklist is a weekly deliverable that provides a concise overview of key health indicators for Inverse Finance products, including FiRM. This checklist includes sections such as Parameter Modeling with Price Impact Data, Collateral Integrity Checkup, and DOLA Health. By setting a regular cadence for updating these models, the checklist allows us to proactively monitor and adjust parameters based on evolving market conditions, ensuring the ongoing robustness of our risk management strategies.
Contract Dependency Index
Designed to map the network of relationships between Inverse Finance’s various smart contracts, data sources, and access control mechanisms. By cataloging each contract’s dependencies—such as oracle price feeds, liquidity pools, or multisig permissions—this framework makes it easier to identify and understand the ripple effects of any change or issue in a single component. For example, if a particular price feed needs updating, the framework helps us quickly determine every contract or product that relies on that feed, ensuring comprehensive impact analysis. Over time, we plan to enrich this dataset and potentially introduce visualization tools to highlight cross-dependencies, enabling the RWG to more effectively manage risks and maintain robust oversight across all Inverse Finance products.
Proof-of-Review (PoR) System
Designed to ensure that all governance proposals—particularly those with significant risk implications—undergo thorough scrutiny before submission; PoRs provide structured sign-offs from key stakeholders, guaranteeing that each change aligns with Inverse Finance’s risk management standards. The current system combines multi-role sign-offs, automated sanity checks, and improved governance. Each proposal must gather specific sign-offs—such as from the RWG, PWG, and contract developers—before it can be submitted. Simultaneously, automated scripts run simulations to catch dangerous configurations (e.g., extreme collateral values or oracle prices), and a revamped UI highlights potential anomalies.