— 01
Market Inputs
Every simulation starts with a live quote and a window of recent daily price history for the ticker. We pull real OHLC bars, normalize them to the current instrument, and feed that into the model before generating any forecast.
Methodology
A probability model, not a crystal ball.
Here's what IndicatorIQ is actually doing under the hood, in plain language.
Beta Notes
Quotes may be delayed during beta. Read the bands as a distribution of possible outcomes, not a recommendation. The model works best on liquid tickers with solid price history.
— 02
Simulation Engine
Under the hood it's a block-bootstrap Monte Carlo — meaning we resample actual historical return blocks rather than assuming prices follow a perfect bell curve. The model estimates recent volatility, generates thousands of forward paths, and surfaces the distribution of where prices could land.
— 03
Sentiment Overlay
When fresh news exists for a ticker, we run it through an LLM extraction pipeline to score relevance, confidence, and novelty. That score gets applied as a short-lived, decaying drift on top of the historical simulation — not a hard override of the whole forecast.
— 04
What You See
The chart shows a median sample path with percentile bands around it. The tighter inner cone is the most likely 25% of outcomes; the outer cone covers a wider probability mass. It's showing you the shape of uncertainty, not a price target.
— 05
What It's Good For
IndicatorIQ works best as a scenario-analysis layer — pressure-testing a setup, comparing tickers side by side, or quickly seeing whether the current distribution looks favorable or stretched.
— 06
What It's Not
It's not financial advice, not a trade signal, and not a guarantee of anything. Volatile or illiquid tickers will produce wider, noisier bands. The model is most reliable on liquid instruments with clean price history.