The numbers you just saw are not answers by themselves. Two investors looking at the same analysis can come to entirely different conclusions — rebalance, add a holding, or simply wait. This page helps you make that decision.
Harry Markowitz's insight that earned him the 1990 Nobel Prize in Economics is surprisingly simple. Don't ask which stock is good; ask how stocks move together. If one falls when another rises, holding both together reduces risk on its own — without giving up return.
Your analysis result dissects your portfolio from exactly this perspective. Volatility, Sharpe ratio, the efficient frontier, correlations — all of them are tools derived from that single line of insight.
On the efficient frontier chart in your analysis page, how close was your portfolio (red dot) to the curve? That distance tells you almost everything.
You're already getting close to the maximum return possible for the risk you carry. Stick to periodic checkups and gentle tweaks rather than big changes.
There is a different weight combination that delivers more return for the same risk. Rebalancing can lift you onto the curve.
Weight tweaking alone may not be enough. Reconsider the holdings themselves, or add an asset class that brings new diversification.
Markowitz analysis won't make decisions for you. It gives you the basis to make them. These are the four most common applications.
Over time, price movements shift weights away from your original intent. Check whether your current portfolio still matches the plan, then use the Equalizer to see how Sharpe responds as you move weights around.
Don't judge by the candidate's standalone return. Look at its correlation with your existing holdings. Two stocks with similar returns can have very different effects on portfolio risk — the one with lower correlation wins.
Your annual volatility tells you the current amplitude your portfolio swings at. Whether to keep, expand, or shrink that amplitude depends on your read of the market. Bullish? Crank volatility up to capture more upside. Cautious? Dial it down for a defensive posture. Markowitz shows the dial; the direction is yours.
Plug the VaR figure into your actual capital. If a ₩100 M portfolio has annual 95% VaR of −16.2%, that means roughly 5 years out of 100 could see losses of at least ₩16.2 M. Whether you can stomach that figure is the starting point of any risk budget.
The better a model is, the more carefully you need to know its limits. Four things to keep in mind before trusting the output blindly.
The result is a hypothesis, not a prophecy. The limits below aren't defects — they are structural properties shared by all quantitative models. The gap between those who know and those who don't is large.
Expected return, volatility, correlations — all of them are estimates from past prices. When market structure shifts (rate-cycle turns, industry paradigm changes), the numbers go stale quickly.
The model assumes returns are bell-shaped, but real markets produce extreme events (2008, March 2020) far more often than theory predicts. This is why VaR is not a reassurance.
A 1-percentage-point change in expected return can shift the optimal weights dramatically. "Reasonable weights under current assumptions" is a more honest phrase than "optimal portfolio."
Commissions, capital-gains tax, slippage, market impact — the model ignores all of them. Frequent rebalancing to chase a tiny Sharpe gain can have those costs swallow the gain entirely.
Markowitz answers "how should I diversify" but not "why does my return look this way." Two portfolios with identical returns can have entirely different sources — broad-market beta, small-cap exposure, value premium. Decomposing the source of return calls for a different model.