Intraday Stock Screening with Amplitude, Large-Order Flow, and Level-One Imbalance

Summary

This stock screen combines three market signals: intraday amplitude above one percent, an afternoon large-order net-inflow condition, and greater displayed buying volume than selling volume at the best bid and ask. The article interprets amplitude as a way to find active shares, large-order flow as a sign of market sentiment, and the level-one volume comparison as a possible indication of buying interest. It supplies formulas and sample data-processing logic, but gives no backtest, performance results, or examples showing that these signals predict returns.

The approach relies on market activity and order-book observations rather than company fundamentals. The author warns that liquidity and sentiment can influence the signals and that financial condition and historical performance are not considered. It recommends testing the screen repeatedly and supplementing it with other technical or company-specific measures. The sample implementation also warrants validation: its use of historical bars and quote data may not faithfully represent a specifically afternoon-based flow signal, and evaluating conditions over multiple observations could produce a different screen from checking a single current snapshot. The proposal is therefore a screening concept whose data timing and signal definitions need clarification before evaluation.

Key ideas

• The screen requires amplitude above one percent, afternoon large-order inflow, and best-bid volume greater than best-ask volume.
• The article treats volatility, large-order activity, and displayed order-book imbalance as indicators of active buying interest.
• It supplies example formulas but no results that establish predictive value.
• Liquidity and changing sentiment can affect these signals, and the screen omits company fundamentals.
• The sample logic's timing and repeated-observation checks should be validated against the intended intraday rules.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.