Portfolio Risk Calculator (Value at Risk)
If you manage an investment portfolio in Vietnam — stocks on the HOSE, local fund certificates, or a mixed book held in Vietnamese dong — sooner or later you ask the question this tool answers: how much could I plausibly lose if things go badly? Value at Risk (VaR) turns that worry into a single number: the maximum loss you would expect not to exceed over a given horizon, at a chosen confidence level, under normal market conditions.
Take the calculator's default setup: a 1,000,000,000 VND portfolio, an annual volatility of 15% (an illustrative assumption), a 95% confidence level and a one-month horizon. The result is roughly 71.230.589 VND, or 7.12% of the portfolio. Read it precisely: "On 95% of months I expect losses no worse than 71.230.589 VND — but in the worst 5% of months, losses can exceed that figure." VaR gives you a threshold, not an absolute ceiling.
Two inputs move this number far more than anything else: volatility and horizon. Higher volatility (concentrated positions, growth stocks) inflates VaR; a longer horizon accumulates risk, but only by the square root of time — not linearly. One caveat governs the whole page: the volatility figures in every example are illustrative assumptions, not measurements of any specific market. Estimate your own portfolio's volatility from its price history or your funds' published data, then enter that into the tool for a realistic answer.
How the calculator computes VaR
The parametric method
This tool uses parametric (variance-covariance) VaR, which assumes returns are normally distributed:
VaR = V × Z × σ × √t
| Symbol | Meaning |
|---|---|
| VaR | Estimated maximum loss (VND) at the chosen confidence level |
| V | Total portfolio value |
| Z | Z-score for the confidence level (95% → 1.645; 99% → 2.326) |
| σ | Annual volatility (standard deviation of returns) as a decimal |
| √t | Time-scaling factor — the square root of the year fraction |
Square-root-of-time scaling
You enter an annual volatility, so the calculator rescales it to shorter horizons by multiplying by √t:
- 1 year: √1 = 1.
- 1 month: √(1/12) ≈ 0.2887.
- 1 day: √(1/252) ≈ 0.0630, using 252 trading days per year.
The crucial consequence is that risk scales with the square root of time, not time itself. One-year VaR is therefore not 12× the one-month figure — it is about √12 ≈ 3.46× larger. This is why a bad month rarely tells you what a bad year looks like.
Confidence level and the Z-score
A 95% confidence level uses Z = 1.645; the stricter 99% level uses Z = 2.326. Moving from 95% to 99% multiplies VaR by 2.326/1.645 ≈ 1.41 — you are asking about a rarer, deeper drawdown, so the threshold rises.
What the model misses — tail risk
Parametric VaR assumes a normal distribution, but real markets have fat tails: large crashes occur more often than the bell curve predicts. So VaR tells you how often losses breach the threshold, never how far past it they go. At 95% confidence, the remaining 5% of periods — the worst ones — can lose dramatically more than the VaR figure. Treat VaR as a normal-conditions warning line, not a worst-case guarantee. To attack the underlying driver, lower portfolio volatility through diversification — see FiMo's portfolio allocation calculator.
Worked example: a 1,000,000,000 VND portfolio at 15%/year volatility (assumed)
Illustrative assumptions: 15% annual volatility, normally distributed returns, no transaction costs. The same portfolio produces very different VaR thresholds depending on the horizon and confidence level.
| Horizon | VaR at 95% (Z = 1.645) | VaR at 99% (Z = 2.326) |
|---|---|---|
| 1 day | 15.543.789 VND (1.55%) | 21.978.634 VND |
| 1 month | 71.230.589 VND (7.12%) | 100.718.754 VND (10.07%) |
| 1 year | 246.750.000 VND (24.68%) | 348.900.000 VND (34.89%) |
Reading the one-month / 95% row: there is a 95% chance this 1B VND portfolio loses no more than 71.230.589 VND over a month. Notice the one-year VaR (246.750.000 VND) is about 3.46× the monthly figure, not 12× — square-root-of-time scaling at work.
Volatility is the lever
Hold the portfolio, horizon (1 year) and confidence (95%) fixed and vary only the assumed volatility:
| Portfolio style (assumed annual volatility) | 1-year VaR at 95% |
|---|---|
| Conservative — bonds/funds (8%) | 131.600.000 VND (13.16%) |
| Balanced (15%) | 246.750.000 VND (24.68%) |
| Growth — equities (30%) | 493.500.000 VND (49.35%) |
The growth book carries nearly double the VaR of the balanced one and almost four times the bond portfolio's — purely because its volatility is higher. That is the quantitative case for diversification and sensible asset allocation: they pull the portfolio's overall volatility down, and VaR falls with it. The 8%, 15% and 30% figures are illustrative; enter your portfolio's actual volatility in the calculator above for a number you can act on.
Frequently asked questions
What is Value at Risk (VaR)?
VaR is the estimated maximum loss a portfolio would not exceed over a given horizon, at a chosen confidence level, under normal market conditions. For a 1,000,000,000 VND portfolio at an illustrative 15%/year volatility, 95% confidence, one-month horizon, VaR ≈ 71.230.589 VND. That means losses stay within this figure on 95% of months — but the worst 5% can run deeper.
What is the VaR formula?
This calculator uses parametric VaR: VaR = V × Z × σ × √t, where V is portfolio value, Z is the confidence-level z-score (95% → 1.645; 99% → 2.326), σ is annual volatility as a decimal, and √t scales it to the horizon. Verifiable: 1,000,000,000 × 1.645 × 0.15 × √(1/12) = 71.230.589 VND for a one-month, 95% estimate.
What's the difference between 95% and 99% confidence?
A higher confidence level asks about a rarer, deeper loss, so the threshold rises. The z-score moves from 1.645 (95%) to 2.326 (99%), multiplying VaR by about 1.41. For a 1B VND portfolio at 15%/year over one month: 71.230.589 VND at 95% versus 100.718.754 VND at 99%. Choose 99% when you want a more conservative read on tail risk.
Why is one-year VaR not 12 times the one-month figure?
Because risk scales with the square root of time, not linearly. The tool converts annual volatility to monthly by multiplying by √(1/12) ≈ 0.2887, so one-year VaR is about √12 ≈ 3.46× the monthly value. Concretely, for a 1B VND portfolio at 15%/year and 95%: monthly VaR is 71.230.589 VND and annual VaR is 246.750.000 VND — exactly the 3.46× ratio.
How does volatility affect portfolio risk?
Volatility is the dominant driver — VaR is directly proportional to it. Same 1B VND portfolio, one-year horizon, 95% confidence: a bond/fund book (assumed 8%/year) shows 131.600.000 VND, a balanced book (15%) shows 246.750.000 VND, and a growth-equity book (30%) reaches 493.500.000 VND — nearly double the balanced case. Diversification lowers overall volatility and pulls VaR down with it.
Is VaR the worst possible loss?
No. VaR is a probability threshold, not a cap. At 95% confidence, 5% of periods still breach the VaR — and can breach it by a large margin. Parametric VaR assumes a normal distribution, while real markets have fat tails (big crashes happen more often than the bell curve implies). Treat VaR as a normal-conditions warning line, and stress-test separately for extreme scenarios.
Where do I get the volatility figure to enter?
Annual volatility (the standard deviation of returns) can be estimated from your portfolio's price history or taken from your funds' published statistics. The 15%, 8% and 30% figures used on this page are illustrative assumptions chosen to make the formula easy to follow. Equity-heavy portfolios are typically more volatile than bond-heavy ones — enter a number that reflects your actual holdings so the VaR is meaningful.