Map · Fabric
A two-dimensional map of the stocks, its day-by-day animation with the selected model's signals, and the market regime.
Market map
What it is: Each point is a stock. Stocks that sit close to each other have residual returns that move together. The lines are the minimum spanning tree, that is, the shortest network connecting each stock to its most similar neighbour.
What data feeds it: Latest analysis run: correlation from the residual returns of the last 500 trading days, distance d = √(2(1−ρ)), classical multidimensional scaling. Colour and size come from the /api/market/frames endpoint: the selected model's backtest signals with a one-day horizon over the last 6 months.
What it shows: In the direction field, the point colour is green for a rise and red for a fall; in the activity field, it is the depth of amber. The size is the magnitude of the value. The edge colour shows the sector. A grey point means no signal that day. The axes have no units; only distances are meaningful.
What can be concluded from it: Stocks of the same sector usually cluster; if a stock is far from its own sector, other dynamics are driving its price. If the animation shows colours spreading from one cluster to a neighbouring cluster, this is visual support for the fabric idea; but it is not measured evidence.
Player
What it is: The play button, date slider and speed setting below the map.
What data feeds it: The list of days in the /api/market/frames response; each frame is one trading day's per-stock values.
What it shows: When playing, the days advance in order; the slider goes to a specific day; the speed is frames per second.
What can be concluded from it: In eventful periods (for example on the days a turbulent regime begins), the whole map turning a single colour shows that the move is market-wide rather than stock-specific.
Regime
What it is: The average correlation between stocks in a 63-day rolling window and the share of the first eigenvalue in total variance; shaded areas are turbulent periods.
What data feeds it: The regime result of the latest analysis run, computed from residual returns. The turbulence threshold is the 70th percentile of the average correlation.
What it shows: When correlation rises, the market moves like a single block; the share of the first eigenvalue measures the same thing in another way.
What can be concluded from it: In turbulent periods diversification does not work and stock-specific signals weaken. C · Change detection raises its alarm threshold in these periods; otherwise every stock would raise an alarm on a crisis day.
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.