> For the complete documentation index, see [llms.txt](https://quandora.gitbook.io/quandora-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://quandora.gitbook.io/quandora-docs/understanding-quandora/glossary.md).

# Glossary

Definitions are intentionally plain English. Ask your AI agent to expand any of these with examples from your own runs.

### Research Objects

**Alpha** — A market signal that appears to predict returns beyond what general market movement (beta) explains. In Quandora, "alpha" and "factor" are used nearly interchangeably.

**Factor** — A measurable market feature turned into a score, e.g. "unusual open-interest change scaled by its volatility." Factors are the research artifact Quandora tests.

**Signal** — The output of a factor: the number per market per bar that says "lean long" or "lean short."

**Task card** — The agent's work order: what to investigate, which data headers are allowed, the forward horizon, and the task status. See [Task Card](/quandora-docs/understanding-quandora/task-card.md).

**plugin.py** — The executable form of a factor that Quandora's server can run. See [plugin.py](/quandora-docs/understanding-quandora/plugin.py.md).

**Factor card** — The structured result report: verdict, evidence, caveats, next experiment. See [Factor Card](/quandora-docs/understanding-quandora/factor-card.md).

**Factor vs strategy vs deployment** — A factor is a research artifact. A strategy is a factor packaged with entry/exit rules, sizing, and risk limits. A deployment is a running instance of a strategy (paper or live).

### Testing Terms

**Backtest** — Replaying history to see how a factor would have scored. Evidence about the past, not a promise about the future.

**In-sample (IS) / Out-of-sample (OOS)** — In-sample is the data a factor was shaped on; out-of-sample is held-out data it has never seen. Strong IS + weak OOS is the classic sign of overfitting.

**Walk-forward** — Repeatedly training on one window and testing on the next, marching through time — a stricter way to catch overfitting.

**Overfitting** — When a factor memorizes historical noise instead of capturing a real pattern. Looks great in backtest, fails live.

**Forward horizon (`fwd_period`)** — How far ahead the factor is judged. Public tasks use 7 daily bars: "does today's score predict the next 7 days?"

**Blindbox** — Quandora's data rule: agents see allowed header names, never the full changing market data. Data binds server-side at evaluation. See [Our Data](/quandora-docs/understanding-quandora/our-data.md).

### Metrics

**Sharpe ratio** — Return per unit of risk. Higher is better; public tasks currently look for ≥ 0.8 cross-sectional Sharpe.

**IC (Information Coefficient)** — Correlation between factor scores and actual forward returns. Positive means the factor points the right way.

**Rank IC** — IC computed on rankings instead of raw values, so outliers don't distort it. The main "does it predict?" metric.

**ICIR** — IC divided by its variability: is the predictive power consistent or streaky?

**IC win rate** — The share of periods where IC was positive.

**Autocorrelation** — How similar the signal is to itself one bar later. Stable signals (high autocorrelation) are cheaper to trade than jittery ones.

**Turnover** — How much the implied portfolio changes between rebalances. High turnover means high trading costs.

**Max drawdown** — The worst peak-to-trough loss over the test. The "see your downside" number.

**Cost viability** — Whether the factor's edge survives realistic trading costs. A great signal that costs more than it earns is not tradeable.

**Regime** — The prevailing market condition (bull / bear / sideways, calm / volatile). Factor cards report the regime mix of the validation window.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://quandora.gitbook.io/quandora-docs/understanding-quandora/glossary.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.