compoundsimPortfolio Monte Carlo Simulator
Mix stocks, bonds, gold and more, find the efficient frontier and test buying on margin, all on real monthly returns.
Simulating a multi-asset portfolio
How the portfolio is simulated
Each of the 1,000 paths comes from a two-regime model, calm and wild, fitted to the shared history of the assets you hold. In the wild regime stocks around the world fall together while bonds and gold often hold up, which is how diversification actually behaved in crashes. The mix is rebalanced to your weights every month. How the model works, and why an asset with a short history widens the band, is explained in the article below.
Reading the efficient frontier
The frontier shows the mixes of your chosen assets with the highest return for each level of risk over the same history, without short selling. Points to the upper left are better. Each asset alone is marked for comparison, and hovering over a point on the line shows its weights. The frontier is fitted to the past: small changes in the period can move it a lot, so treat it as a map of past trade-offs, not a forecast.
Buying on margin
With leverage above 1, the starting sum and every contribution buy that multiple of your own money, with the rest borrowed at the borrowing rate. Interest is added to the loan each month. If equity falls below the maintenance margin at a month end, holdings are sold to bring it back, as a broker's automatic liquidation would. A loan in another currency also moves with its exchange rate against the dollar, so a rise in that currency adds to the debt.
Reading the forced-sale risk
The risk chart replays the same draws at every leverage from 1.0 to 3.0 and shows how often a forced sale happens and how often you end below the money you paid in. Clicking a ratio applies it. Prices are checked at month ends only, so falls that recover within a month never trigger a sale here: real risk is higher than shown.
About the data
The model is fitted to monthly total returns over the history you choose, every asset over the months it covers. Bitcoin's history starts in 2014 and US dividend stocks in 2011, so their behaviour in the wild regime rests on few months and is partly borrowed from how the other assets moved in those months; the band is wider for them. The stats line shows the months every held asset covers. Past returns do not predict future ones, and none of this is investment advice.