Monte Carlo Forecasting with Normal Returns or Bootstrap Resampling
Summary
This indicator generates multiple simulated paths either for price or for a user-supplied strategy return series. Users can choose a normal model, which samples from the historical mean and standard deviation, or bootstrap resampling, which draws observed returns. For price projections, historical log returns are collected from a configurable lookback and applied multiplicatively; for strategy results, pasted returns are accumulated additively.
The output can show projected paths, best and worst paths, endpoint distributions, or a selected cumulative-probability range. The script also offers seed and forecast controls, simulation count, and histogram binning choices. These are scenario illustrations based on the chosen input data and distribution assumption, not evidence that future outcomes will follow the simulated paths. Normal sampling may miss skew and tail behavior, while bootstrap results depend on the sample representing future conditions; the document provides no validation results or guidance on dependence between returns.
Key ideas
- The indicator simulates price paths from historical log returns or strategy paths from pasted returns.
- Normal mode uses the sample average and standard deviation, while bootstrap mode resamples observed returns.
- Users can inspect individual paths, extremes, endpoint distributions, or a cumulative-probability region.
- Simulation output is conditional on the selected return history and sampling assumptions, so it does not establish forecast reliability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.