Stock 1 day prediction under a dollar

Screening Filters

1. Price: 0.10 – 1.00

   • Purpose: Focus on stocks “under a dollar,” while excluding the most extreme micro-penny names.
   • Rationale:
     • The user wants prediction for sub‑$1 stocks. Setting the upper limit at 1.00 aligns directly with that.
     • The 0.10 lower bound removes ultra-illiquid, highly erratic tickers (e.g., $0.001) where price moves and models are less meaningful and spreads are huge.

2. Monthly Average Dollar Volume: ≥ 300,000

   • Purpose: Ensure there is enough trading activity (liquidity) for the predictions to be usable.
   • Rationale:
     • Illiquid penny stocks can show random, “one-trade” price jumps that don’t reflect real investor behavior.
     • A minimum dollar volume filter improves the reliability of any 1‑day prediction model, and makes it more feasible to enter/exit positions near the quoted prices.

3. RSI Category: Moderate or Overbought

   • Purpose: Focus on stocks with identifiable short-term momentum conditions that are relevant for 1‑day performance.
   • Rationale:
     • Moderate RSI: Not stretched in either direction, leaving room for a move if other signals are positive.
     • Overbought RSI: Indicates recent strength / upward momentum; many short-term models treat positive momentum as a factor for near-term gains (with the caveat of pullback risk).
     • Restricting to these RSI zones targets names where short-term technical conditions give the model more structure to work from, rather than completely directionless charts.

4. One-Day Rise Probability: ≥ 55%

   • Purpose: Select stocks where the model indicates better-than-random odds of rising tomorrow.
   • Rationale:
     • A 50% probability is essentially a coin flip.
     • A minimum of 55% focuses on situations where the predictive model sees a slight statistical edge toward an up day, which is exactly what the user is asking about: 1‑day performance expectations.

5. One-Day Predicted Return: ≥ 3%

   • Purpose: Filter for stocks where the size of the predicted move is meaningful, not just the probability.
   • Rationale:
     • Even if the rise probability is above 50%, a predicted move of, say, +0.5% might not be worth trading.
     • A 3% minimum targets names where the model expects both a positive direction and a non-trivial magnitude of 1‑day return, making the predictions more actionable.

Why Results Match the User’s Request

• The price filter confines the universe strictly to sub‑$1 stocks, precisely matching the “priced under a dollar” requirement.
• The liquidity filter (dollar volume) ensures those predictions are based on meaningful trading activity, not random ticks.
• The RSI filter narrows the list to stocks with clearer short-term technical setups, which are more relevant for 1‑day prediction models.
• The one-day rise probability and predicted return filters directly operationalize the request to “predict the 1-day performance” by selecting only those sub‑$1 stocks where the model forecasts a better-than-random chance of going up and a reasonable expected gain.