Quantus Investing
AI-native market research that turns a ticker into an explainable signal, a 7-section report, and an accuracy trail you can verify before acting.
ML/Statistical models
Report sections
Backtested equities
Live research preview
Search to signal in one place
Coverage
Equities, ETFs, crypto, and commodities across 10 sector groupings.
Data sources
SEC EDGAR, FRED, and CBOE as primary sources, plus real-time sentiment from Grok/X, Reddit, and NewsAPI.
Validation
Walk-forward backtesting since 2018 with a public accuracy dashboard segmented by signal, regime, sector, and market cap.
Built on 12 models, verified through public backtesting
Every capability below is backed by documented methodology, tiered data sourcing, and walk-forward validation across 400+ equities since 2018.
12-Model Ensemble
LSTM (45%), Prophet (35%), and ARIMA (20%) form the core forecasting stack, with additional models covering sentiment, risk, options, factor analysis, and reinforcement learning.
7-Section Research Reports
Each report combines executive summary, ensemble forecasts, regime detection, risk framing, strategy guidance, and on-demand deep dives in a single research workflow.
Regime-Aware Signals
Signals are generated after classifying the market into one of five regimes so model weighting, strategy framing, and position sizing change with conditions.
Sector Packs
Pre-generated teardowns for the top 20 tickers across 10 sectors let users move beyond one-off reports into grouped market context.
Accuracy Dashboard
Backtested outcomes are segmented by signal type, engine version, regime, sector, and market cap so performance can be checked in context.
Confidence Scoring
A 0-100 composite score combines Momentum, Sentiment, Regime Alignment, Model Ensemble Agreement, Alternative Data, Macro Context, and Data Quality.
Launch a Quantus workflow from the BI Solutions product shell
The BI Solutions site acts as the product shell and discovery layer, while the Quantus workspace handles report generation, sector packs, archive comparisons, and deeper signal review at `/quantus/workspace/`.
Workspace URL
https://www.bisolutions.group/BI-Solutions-Advanced-Analytics-AI-Consulting/quantus/workspace/
Suggested research flow
- 1
Search any asset
Enter a ticker, company, crypto pair, or commodity. Cached reports load instantly when coverage exists, or generate a fresh report in seconds.
- 2
Read the full research report
Review a 7-section report with executive summary, ensemble forecasts, regime classification, risk framing, strategy recommendations, and deep dives.
- 3
Track, compare, and act
Use watchlists, signal-change alerts, and archive comparisons to follow how a name changes over time.
Quantus Investing FAQ
What data sources does Quantus use?
Quantus uses a tiered data source system. Tier A includes SEC EDGAR filings, FRED economic data, and CBOE volatility indices. Tier B includes Yahoo Finance and Financial Modeling Prep for price data. Tier C includes Grok/X, Reddit, and NewsAPI for sentiment - these are advisory inputs and never override Tier A sources.
How many models are in the ensemble?
Meridian v2.4 runs 12 ML and statistical models. The three core forecasting models are LSTM (45% weight), Prophet (35%), and ARIMA (20%). The remaining nine cover sentiment analysis, mean reversion, portfolio optimization, SHAP explainability, high-frequency signals, VaR risk, options pricing, reinforcement learning, and factor analysis.
How is accuracy measured?
The Accuracy Dashboard tracks resolved outcomes from persisted report snapshots. Results are segmented by signal type, engine version, market regime, sector, and market cap so users can evaluate Quantus performance in conditions relevant to their strategy.
What does the confidence score mean?
The confidence score (0-100) is a weighted composite of seven sub-signals: Momentum, Sentiment, Regime Alignment, Model Ensemble Agreement, Alternative Data, Macro Context, and Data Quality. A score with low data quality should be interpreted differently than the same score with high data quality.
What asset classes are covered?
Equities, ETFs, cryptocurrencies, and commodities. Sector Packs provide pre-generated coverage for the top 20 tickers across 10 sectors, refreshed every 96 hours. Individual reports can be generated for any supported ticker on demand.
Is this financial advice?
No. Quantus Investing is a research and analysis tool. It supports investment research workflows but does not provide financial advice, manage portfolios, or execute trades. All signals, scores, and recommendations are informational. Final interpretation and investment decisions remain with the user.
Can I compare old reports to current ones?
Yes. The Archive stores every report snapshot with a unique report ID. You can load any historical snapshot and run a side-by-side diff against the current report to see how signals, regime classification, and confidence scores have changed over time.