TERMINAL / GUIDE
Monte Carlo
Explore distributions of paths sampled from observed historical return blocks.
13 / 14Monte Carlo
Interactive interface · ArcDelta dataOpening the workspace…
Start here
- 01Select asset, interval and available historical window.
- 02Set the horizon and number of paths, then run the calculation; the method panel reports block length.
- 03Read quantile bands, terminal distribution and validation status.
Inputs
Available ArcDelta historical closes for the selected asset and interval.
Requires Labs access; the statistical interpretation is public.
General limitations
Historical samples and simulated outcomes do not establish a price-direction forecast. Small samples, overlapping events and changing regimes limit inference.
- Calculation
- The live workspace samples contiguous blocks of historical log returns and compounds them from the starting price. A block keeps local dependence within its sampled segment.
P(t + 1) = P(t) × exp(sampled log-return) - Reading & controls
- Changing the historical window, horizon or block length changes the assumed scenario distribution. Repeat with the same settings to compare support.
- Limitations
- Resampling does not create new market observations and cannot guarantee the current regime matches history.
- Calculation
- At each step the path sample is summarized by its quantiles. The terminal distribution describes the paths' ending values.
- Reading & controls
- Inspect p05, p25, p50, p75 and p95 with the horizon. Distinguish a pointwise band from the probability that an entire path stays inside it.
- Limitations
- Simulated quantiles are not validated forecast confidence intervals. Coverage must be measured on new real outcomes.