Today
The starting screen for daily use. It summarises the signals produced by the latest pipeline run and gathers the stocks that need attention in two lists. There are no parameters; only the horizon is chosen.
Status strip
What it is: The summary line at the top of the screen: the day the signals belong to, the tier (Live, Replay or Backtest), the kind of sector factor, the day's counts of the four consensus states, and the horizon selector.
What data feeds it: The /api/today endpoint. The latest LIVE day is looked up in the signal_daily table; if there is no live record at all, the latest backtest day is used. The state counts are counted from the state field of that day's D · Consensus rows.
What it shows: It shows which day's signals you are looking at and how the consensus distributed the stocks that day among the "Strong signal", "Event, direction unclear", "Weak" and "None" states.
What can be concluded from it: If the tier says "Backtest", the signals on the screen are the model applied to the past and are not for today's decision; run the pipeline first. It is normal for the number of strong signals to be very small; on most days no stock has both an event and agreement on direction.
Strong signal cards
What it is: Cards for the stocks on which C · Change detection says "something is happening" and the two direction models (A · Graph diffusion and B · Cross-sectional ranking) point the same way. Sorted by confidence, largest first.
What data feeds it: /api/today, D · Consensus, the selected horizon, state = GUCLU_SINYAL. The mini chart is the last 30 adjusted closes in the price_daily table; the daily change is computed from the last two closes.
What it shows: The arrow is the sign of the average of the A and B directions. Confidence is the product of the activity of C · Change detection and the A–B agreement on direction; that is, it measures the strength of both the event and the agreement together.
What can be concluded from it: This list means "something unusual is happening in this stock and two independent methods say the same direction"; it is not a guarantee of return. If the Performance screen says the consensus does not beat its components, these cards should not be used for decisions. Click a card to look at the outcome of similar past signals on the tracking chart.
Event, direction unclear
What it is: The list of stocks on which C · Change detection raised an alarm but A and B point in opposite directions. Sorted by activity.
What data feeds it: /api/today, D · Consensus, the selected horizon, state = OLAY_VAR_YON_BELIRSIZ. The activity value is the "probability that a change has occurred in the last 3 days".
What it shows: It shows the stocks in which an unusual break in volume, volatility or residual return was caught, but on whose direction the models do not agree.
What can be concluded from it: These are the stocks for which you should check the news, KAP disclosures or the earnings calendar. The system gives no information on direction; in this situation waiting is usually the only honest response.
All signals table
What it is: A table, collapsed by default, listing row by row the signals of all models at all horizons for that day.
What data feeds it: The "all" list in the /api/today response; all rows of the signal_daily table for that day and tier.
What it shows: Each row has the stock, model, horizon, direction, activity, confidence and consensus state. It can be sorted and filtered by column; clicking a row opens the stock screen.
What can be concluded from it: Used to see why the models disagree on a stock: for example, if A · Graph diffusion says up and B · Cross-sectional ranking says down, D · Consensus becomes "direction unclear". It is the place to examine consistency between models one by one.
The investment information, comments and recommendations given here are not within the scope of investment advisory services. Investment advisory services are provided under an investment advisory agreement to be signed between a client and brokerage houses, portfolio management companies, or banks that do not accept deposits. The signals here are produced from historical data with statistical models, are shown the same to everyone and are not personalised; they may not suit your financial situation or your risk and return preferences. Therefore, making investment decisions based solely on the information given here may not produce results that meet your expectations.