Prediction-Interval Fan Chart¶
Uncertainty visualization — instantiates Stochastic Process Envelope Modeling
Displays a forecast as a widening fan of probability bands over the horizon, showing how the range of plausible outcomes grows the further ahead you look.
Prediction-Interval Fan Chart renders a forecast's uncertainty as a set of nested shaded bands — a central estimate wrapped in, say, 50%, 80%, and 95% intervals — that widen as the horizon extends. Its defining move is that it summarizes the process as marginal quantile bands at each point in time: at every future step it shows the range the outcome is likely to fall in, and stacks those ranges into a fan. Onto that envelope it lays the thresholds a decision actually turns on, so a reader sees not just "uncertainty grows" but "here is the probability of crossing the line that matters." It is a communication instrument first — its job is to make growing uncertainty and threshold risk legible at a glance, without demanding the reader run the model.
Example¶
A central bank publishes an inflation forecast in the style the Bank of England made famous with its fan chart. The central projection runs out three years, wrapped in bands that widen steadily, so next quarter's figure is shown as fairly tight while the two-year-ahead figure sprawls across a wide range. A horizontal line marks the 2% target, and the shading makes visible the probability that inflation breaches, say, 4%. Building it means taking the forecast's predictive distribution at each quarter, extracting the quantiles, shading the nested intervals darkest at the center, and annotating the target line with its exceedance probability. The effect on the reader is immediate: attention shifts from arguing about a single headline number to reasoning about how likely inflation is to overshoot, and how much less certain the distant horizon is than the near one.
How it works¶
- Take the per-horizon predictive distribution. At each future step, obtain the forecast's distribution — analytically or from an ensemble of simulated paths.
- Extract the quantiles. Pull the median and the interval edges (e.g., 10th/90th, 2.5th/97.5th) for each step.
- Shade the nested bands. Draw the intervals as concentric shaded regions, darkest at the center, widening with horizon.
- Overlay the decision thresholds. Add the lines that matter and annotate the probability of crossing them.
- Flag the marginal caveat. The bands describe each step in isolation; they say nothing about whether a high early value implies a high later one.
Tuning parameters¶
- Which intervals to show — 50/80/95, or fewer; more bands convey detail but clutter the picture.
- Symmetric vs skewed bands — allowing asymmetric intervals when risks lean one way rather than forcing a symmetric fan.
- Horizon length — how far out to project; further horizons show more uncertainty but less reliably.
- Threshold overlays — which decision lines to draw and how exceedance probability is computed.
- Central tendency shown — median, mean, or mode as the fan's spine, which changes what "the forecast" reads as.
When it helps, and when it misleads¶
Its strength is communication: it conveys growing uncertainty and threshold risk to non-technical audiences better than any table, and it reframes a decision around the probability of crossing a line rather than a false-precise point estimate.
Its failure mode is that it hides path dependence. Because the bands are marginal per period, a fan chart[n1] does not show whether an outcome high early stays high, yet readers routinely misread the shaded region as a bundle of smooth paths any of which is equally likely. Narrow bands built from an underestimated model uncertainty compound the danger by projecting false confidence. The classic misuse is treating any tidy curve drawn inside the fan as a plausible scenario, ignoring that early and late errors are correlated. The guarding discipline is to state plainly that the bands are marginal and to pair the chart with a handful of actual sample paths whenever the shape of a trajectory — not just its range at each step — drives the decision.
How it implements the components¶
ensemble_path_envelope— the shaded, widening quantile bands are the rendered envelope of plausible outcomes across the horizon.risk_threshold_policy— the overlaid decision lines and their annotated exceedance probabilities turn the envelope into an actionable alert about crossing a limit.
It shows the envelope but not the paths that make it: scenario_path_ensemble and path_trace_archive — generating and storing the individual whole trajectories — belong to Trajectory Ensemble Simulation, which can supply the very quantiles this chart summarizes.
Related¶
- Instantiates: Stochastic Process Envelope Modeling — the fan chart is the display layer that makes the process envelope readable and decision-linked.
- Consumes: Trajectory Ensemble Simulation can supply the simulated paths from which the quantile bands are computed.
- Sibling mechanisms: Trajectory Ensemble Simulation · Markov Chain Model · State-Transition Kernel · Poisson Event Model · Sequential Filter Update · Innovation Residual Monitor · Drift Recalibration Loop · Stationarity Check · Stochastic-Process Diagram
Editorial Notes¶
Form Classification¶
Form family: Interface, Display & Cue
Rationale: Prediction-Interval Fan Chart operates as a user-facing prompt, display, template, or perceptual cue that shapes attention and action at the point of use because it displays a forecast as a widening fan of probability bands over the horizon, showing how the range of plausible outcomes grows the further ahead you look.
Independent corroboration: The frozen evidence defines Prediction-Interval Fan Chart as 'Displays a forecast as a widening fan of probability bands over the horizon, showing how the range of plausible outcomes grows the further ahead you look', so its operative form is Interface, Display & Cue.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Economics & Finance
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Fan charts were institutionalized in macroeconomic forecasting to communicate widening uncertainty over time.
Related originating lineages:
- Statistics & Experimental Design — Statistics supplies prediction intervals and probability-band construction.
Review resolution: Light authoritative-source research resolves the primary-origin disagreement in favor of economics finance. Bank of England: The Inflation Report Projections—Understanding the Fan Chart directly documents the defining practice or theory described in the selected origin rationale. Other domains are retained only where the blind reviews identify material co-development or translation; broad application is recorded separately as domain_reach=multi_domain, while origin_mode=cross_disciplinary_synthesis describes the relationship among origin lineages.
Attribution caveat: The boundary with statistics experimental design is substantive because that tradition materially developed or translated part of the mechanism; the cited provenance places the defining form in economics finance.
Review outcome: Researched adjudication after independent review; high confidence.
Sources consulted:
Notes¶
[n1] The fan chart, popularized by the Bank of England's Inflation Report, shades predictive intervals that widen with the forecast horizon; a well-known pitfall is that the bands are marginal at each period and therefore do not depict the correlation between one horizon and the next. ↩