> ## Documentation Index
> Fetch the complete documentation index at: https://methodscenter.mintlify.app/llms.txt
> Use this file to discover all available pages before exploring further.

# API Reference

> Interactive documentation for the Luna Modelling API.

## Welcome to the Luna Modelling API

The Luna Modelling API provides advanced statistical modeling and filtering services designed for educational research and student performance analysis. Built with robust quota management and persistent data storage, our API enables researchers and institutions to process time series data efficiently.

<Note>
  Explore the endpoints below using the live OpenAPI-powered playground.
  Switch between request examples to understand stateless and stateful Kalman
  filter runs.
</Note>

## Getting Started

To start using the Luna Modelling API:

1. **Contact us** to obtain your API key and account ID
2. **Include your API key** in the `X-API-Key` header with every request
3. **Monitor your quota** using the `/account/quota` endpoint
4. **Process your data** with our Kalman filtering service

### Base URL

```
Production: https://api.luna-modelling.com/v1
Development: http://localhost:8000/v1
```

## Specification

<Card title="Luna Modelling API" icon="code" href="/api-reference/openapi.json">
  Download the OpenAPI specification that powers this reference.
</Card>

## Authentication

All endpoints require an API key via the `X-API-Key` header. The server uses the
key to validate access, enforce quotas, and bind the request to the supplied
`account_id` path parameter when relevant.

```http theme={null}
X-API-Key: <your-api-key>
```

## Available Endpoints

### Account Management

* **`GET /account/quota`** - Check your remaining quota before issuing modelling requests. Use this to monitor usage and plan your API calls accordingly.

### Data Processing

* **`POST /{account_id}/kalman`** - Run Kalman filtering on one or more time series, with optional persistence. The Rauch-Tung-Striebel smoother reduces noise and provides smoothed predictions for student performance data.

## Use Cases

The Luna Modelling API is designed for:

* **Student Performance Prediction** - Filter noisy grade data to identify true performance trends
* **Early Warning Systems** - Detect students at risk of dropout through statistical analysis
* **Longitudinal Studies** - Accumulate and analyze student data over multiple semesters
* **Research Projects** - Process educational data with state-of-the-art filtering techniques

Use the sidebar to jump straight into interactive examples for each endpoint.