# AI Strategy Parameters & Tuning

## 1. **Strategy Template: HFRR-CLMM**

### Name & Type

HFRR-CLMM: High-Frequency Range Rebalancing for Concentrated Liquidity Market Making. It's a dynamic strategy tailored for modern CLMM (Concentrated Liquidity Market Maker) AMMs like Cetus and Uniswap V3, which focuses on ultra-narrow price bands with frequent rebalancing to maximize fee capture while mitigating impermanent loss.

### Target Pairs

Initially deployed on the SUI/USDC trading pair. Future deployments may include high-volume, high-volatility pairs (e.g., DEEP/SUI, CETUS/USDC).

### AMM Integration (e.g., Cetus on Sui)

Deployed on Cetus AMM, a Uniswap V3-style CLMM built for the Sui blockchain. Integrates with Cetus SDK for price feeds, tick data, liquidity depth, and gas cost estimation. Smart contracts interact directly with pool objects to add/remove liquidity within selected ticks.

## **2. Key Parameters**

### ΔP\_range (Price Range Width)

This defines the tightness of the liquidity band relative to the current price P\_c. Narrower bands = higher fee capture, higher rebalance frequency, higher gas costs.

* Example: ΔP\_range = ±0.10% → liquidity placed within 0.1% of P\_c.

### δ\_trigger (Rebalance Threshold)

The % deviation of the current price from the range boundary that triggers a rebalance. It can be defined in % or tick units.

* Example: If P\_c crosses P\_min or P\_max by 0.02%, and δ\_trigger = 0.01% → rebalance is initiated.

### T\_hold\_min (Minimum Hold Time)

Minimum time (in seconds/minutes) before a position can be rebalanced again. It prevents over-trading and protects against excessive gas usage, which ensures the strategy commits capital for a minimum interval to collect fees.

### σ\_threshold (Volatility Cutoff)

It determines if market volatility is too high for narrow-band liquidity provision. If realized or implied volatility exceeds σ\_threshold, the strategy may pause or widen bands based on the rolling standard deviation of price (σ\_market).

### C\_gas\_profit\_ratio (Gas-Profit Threshold)

Strategy only rebalances if the expected fee gain exceeds gas cost by a set ratio. Helps maintain profitability in volatile or expensive gas conditions.

* Formula:&#x20;

```
Expected Fee / Estimated Gas Cost > C_gas_profit_ratio
```

### Event Window Logic

Defines time windows to pause trading around economic events (e.g., CPI reports, FOMC). Window = pre-event buffer + post-event cooldown. It helps prevent execution during unstable liquidity or low predictability.

## **3. Parameter Optimization (Backtest Results)**

### Summary of How Parameters Are Tuned

Parameters are optimized using historical tick-level data from Cetus for 5+ months. Simulations assess performance across multiple volatility regimes and fee environments. Objectives include maximizing net APY, reducing IL, and minimizing gas cost per rebalance.

### Default vs Optimized Values

| Parameter             | Default | Optimized Range (example) |
| --------------------- | ------- | ------------------------- |
| ΔP\_range             | ±0.15%  | ±0.05% – ±0.25%           |
| δ\_trigger            | 0.01%   | 0.005% – 0.015%           |
| T\_hold\_min          | 15 min  | 5 – 30 min                |
| σ\_threshold          | 2.5%    | 1.5% – 3.0%               |
| C\_gas\_profit\_ratio | 1.5     | 1.2 – 2.0                 |

🚧 Placeholder: Final tuning values pending backtest report from VAR Meta team.

### Strategy Switching Logic (Agentic vs Manual)

* Agentic: AI agent monitors volatility, volume, fee yield, and pool depth to autonomously switch parameters or strategies.
* Manual: DevOps or strategy team may override via governance hooks or emergency admin tools.

The long-term goal is full autonomy with periodic human oversight for safety checks.

<br>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.nodo.xyz/public/user-policy/ai-agent-vault-details/ai-strategy-parameters-and-tuning.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.