FAQ

1. What is ZOIA?

ZOIA is a deep learning–based AI system that generates next-trading-day high and low price predictions for selected U.S. stocks.

It analyzes large-scale market time-series data to produce trade-grade outputs for institutional and professional use.

ZOIA does not execute trades, provide investment advice, or generate trading signals.

2. How is ZOIA applied within institutional workflows?

ZOIA analyzes large-scale historical market time-series data to generate next-trading-day high and low price predictions. These outputs are delivered to institutional and professional users and integrated into client-defined decision and downstream execution workflows.

3. What makes ZOIA different?

ZOIA is designed around discrete, next-trading-day high and low price predictions, provided as a standardized daily deliverable for a defined universe of U.S. stocks.

By focusing on clearly defined price predictions rather than directional instructions, each output can be assessed using objective quantitative measures such as deviation from realized market prices and relevance within client-defined trading or risk frameworks.

This design emphasizes auditability, traceability, and reproducibility in scalable production workflows, enabling consistent integration into institution-scale decision, risk, and execution processes under client-defined controls and limits.

4. How does ZOIA leverage deep learning and AI?

ZOIA uses deep learning to model financial time-series data and generate next-trading-day high and low price predictions for selected U.S. stocks.

The system applies neural network–based modeling to capture non-linear dynamics in historical market data and produces standardized predictive outputs for institutional and professional use.

5. How often are predictions generated?

Prediction outputs are generated daily for the covered universe.

Model improvements are introduced through ongoing R&D and production maintenance, without a fixed public release cycle.

6. How are predictions evaluated?

ZOIA evaluates its predictions by comparing predicted next-day high and low prices against realized market high and low prices using predefined quantitative error and deviation metrics.

Historical prediction and evaluation records are maintained for monitoring and audit purposes, and are available to authorized users.