Monte Carlo Investment Simulator

How Monte Carlo simulation works

Monte Carlo simulation generates thousands of random scenarios, each with random annual returns drawn from a normal distribution around your expected return, with the spread determined by volatility. Each scenario compounds month by month, adding contributions along the way. After all scenarios finish, you sort the final portfolio values to extract percentiles: the 10th percentile (worst 10%), median (middle), and 90th percentile (best 10%).

This approach naturally captures the real risk of markets: some years are great, some are disastrous, and the order is unpredictable. By running thousands of scenarios, Monte Carlo shows you the statistical likelihood of different outcomes.

Interpreting the percentiles

• 10th percentile: Only 10% of scenarios end up lower. This is downside risk - the outcome in a bad-luck sequence of poor market years. Use this to check if your floor is acceptable.
• 50th percentile (median): The middle outcome; half the scenarios do better, half worse. A reasonable target for planning.
• 90th percentile: Only 10% of scenarios end up higher. This is upside potential - the outcome if you get lucky. It is tempting to plan around this, but dangerous.

Many financial advisors suggest aiming for a "success rate" of 85-95%, meaning your goal is safe in 85-95% of scenarios. This typically corresponds to the 25th-50th percentile outcome, depending on your risk tolerance.

Critical assumptions

Monte Carlo is only as good as your inputs:

1. Expected return: Historical equities average 10-12% in India (nominal), global ~8-10%, bonds ~6-8%. Individual years are wild; decades are predictable. Reduce your estimate by 1-2% to account for taxes, fees, and the humble possibility that you are wrong. 2. Volatility: Historical equity volatility is 15-20%; bonds 5-8%; 70/30 portfolios ~10-12%. Real volatility is lumpy (some years plus or minus 40%, some plus or minus 5%), not smooth. Higher volatility means wider percentile spreads. 3. Time horizon: Monte Carlo requires years to work. A 5-year plan can still drop 40% mid-way; a 20-year plan smooths that out. This is why starting early matters. 4. Rebalancing: The simulation assumes consistent return and volatility. In reality, as markets rise, equities become a larger share of your portfolio, increasing volatility (unless you rebalance). Rebalancing annually or semi-annually reduces actual volatility.

Common mistakes

Underestimating volatility - Many investors assume returns are steady; they are not. A "12% return" with "8% volatility" means you could see +20% or +4% in any given year. Overconfident assumptions lead to overconfident plans.

Planning for the 90th percentile - Tempting, but risky. If 90% of scenarios need to succeed for your plan to work, one unlucky sequence of market returns ruins it. Aim for the 75th percentile at best; the 25th-50th is conservative.

Ignoring sequence risk - Bad returns early hurt more than bad returns late, because you have fewer years to recover. Monte Carlo automatically models this: some simulations get lucky (good markets early), others unlucky (crashes early). This is why retirement planning is about the full distribution, not just the average.

Forgetting taxes and fees - If you assume 12% returns, that is usually pre-tax, pre-fee. Net returns might be 8-10% after taxes (depending on the instrument) and 0.5-1.5% annual fees (mutual fund / advisor). Use net returns in the simulation.

Not rebalancing - If you start 70% stocks / 30% bonds and let it drift to 80% / 20%, your volatility rises but your expected return does not keep pace. Annual rebalancing keeps volatility stable and forces you to sell winners and buy losers - a sensible discipline.

Using this for retirement planning

Set your goal (e.g., INR 1 crore by age 60), adjust monthly contributions until the 25th percentile outcome just meets it, then check that you can actually afford that monthly amount. If not, retire later or adjust the goal. Run the simulation out to age 100 to make sure your corpus lasts. A common rule: you can safely withdraw 3-4% of your corpus annually without running out in 30+ years, even accounting for failed simulations. So INR 1 crore supports INR 30-40k/month of spending in retirement.