Jump Models Server

​

Overview

​

Connection

{
  "mcpServers": {
    "jump-models": {
      "url": "https://mcp-production-11da.up.railway.app/mcp/jump-models"
    }
  }
}

​

Transport

​

Best-Fit Workflows

• Fit JumpModel or SparseJumpModel regimes from numeric feature matrices.
• Run online state prediction and state-probability prediction.
• Backtest regime state exposure into strategy rows and metrics.
• Chain CSV, table, result, and artifact outputs into model tools.
• Use the included Nasdaq-100 example data path.

​

Recommended Tools

• get-data
• table_dropna -> table_to_ndarray
• jump-model-create -> jump-model-fit -> jump-model-predict-online
• sparse-jump-model-create -> sparse-jump-model-fit -> sparse-jump-model-predict-online
• jump-model-backtest
• save_tool_result

​

Tools

​

Examples

​

Run a Simple State Backtest

{
  "request": {
    "rows": [
      {
        "timestamp": "2024-01-02",
        "ret": 0.01
      },
      {
        "timestamp": "2024-01-03",
        "ret": -0.02
      },
      {
        "timestamp": "2024-01-04",
        "ret": 0.005
      }
    ],
    "states": [
      0,
      1,
      0
    ],
    "state_exposure": {
      "0": 1.0,
      "1": 0.0
    },
    "transaction_cost_bps": 1.0
  }
}

​

Create a JumpModel

{
  "request": {
    "init_kwargs": {
      "n_components": 2,
      "n_init": 3,
      "jump_penalty": 0.0
    }
  }
}

​

Fit a Prepared Matrix

{
  "request": {
    "handle": "<handle-from-create>",
    "X": {
      "result_id": "<matrix-result-id>"
    },
    "ret_ser": {
      "result_id": "<return-vector-result-id>"
    },
    "sort_by": "cumret"
  }
}

​

n8n Workflow

1. Trigger on a schedule or webhook and set tickers, start_date, end_date, jump_penalty, n_components, and n_init.
2. Fetch market data through the Dataloader MCP endpoint, then call result_get if the extraction returns result_id.
3. Build numeric features in a Code node or with table helper tools.
4. Call jump-model-create, then jump-model-fit with X, ret_ser, and sort_by: "cumret".
5. Call jump-model-predict-proba-online and jump-model-predict.
6. Resolve both prediction result_id values with result_get.
7. Format the final regime report in n8n and respond through the webhook or delivery nodes.

• Regime Model JSON - public regime-report webhook at POST https://n8n.limex.pro/webhook/9f2988bf-7247-4c07-95c7-9e37188e93c9. The template removes the scheduled email-delivery branch; add your own delivery nodes after import.

​

Notes

• table_to_ndarray outputs can be passed directly to fit and predict tools as X={"result_id":"<matrix_result_id>"}.
• Use table_dropna or table_impute before table_to_ndarray when feature columns can contain blanks or nulls.
• jump-model-backtest is the native path for equity, strategy return, drawdown, Sharpe, and turnover metrics.
• The hosted endpoint is intentionally no-bearer; omit the headers block for this server.

​

Other Servers