This Sophon-3 AI model flow starts with the DIJA-30 as the asset universe, then chains together two AI model nodes: Sophon-3 Screener and Sophon-3 Forecast.
The idea is simple: first narrow the universe to more interesting candidates, then forecast their near-term return potential before sending the results into portfolio construction. This creates a practical workflow where screening, forecasting, and position sizing work together instead of as separate steps.
In this template, Sophon-3 Screener uses data types including Dividend Yield and Total Assets. Dividend Yield measures the company’s annual dividend relative to its current stock price; it provides a quick read on income return and whether a stock’s payout is meaningfully contributing to total return. Total Assets is the value of everything the company owns on its balance sheet; it helps describe the company’s scale and asset base that supports operations.
Together, these inputs give the screener both an income lens (payout versus price) and a balance-sheet scale lens (how large the asset base is).
Sophon-3 Screener is an AI model node that continually learns from new market dynamics instead of relying on fixed screening rules. Its role is to filter the DIJA-30 toward stocks with stronger potential versus the benchmark.
The selected names are then passed to Sophon-3 Forecast, another AI model node, built for continual deep learning forecasting. Sophon-3 Forecast focuses more directly on price dynamics and estimates near-term return direction and magnitude.
Chaining the two nodes is meaningful because they solve different parts of the problem. Sophon-3 Screener helps reduce noise by finding better candidates, while Sophon-3 Forecast complements it by analyzing price behavior and short-term timing. Those forecasts then flow into portfolio construction, where stronger conviction can translate into larger position sizes.
