On this page
What this lesson covers
A confidence interval estimates a population parameter with a range constructed at a chosen confidence level.
Build a large-sample interval from a sample mean and standard error. Distinguish its two-sided critical value from the one-sided loss threshold used for parametric value at risk.
How do you calculate confidence intervals and VaR?
For a large sample, add and subtract the critical Z-value times the standard error; use the sample standard deviation when the population value is unknown. Parametric VaR instead uses a one-sided Z-value, normally distributed returns and a zero mean.
Confidence interval = sample mean ± critical Z-value × standard error
- Sample mean
- Two-sided critical Z-value
- Population standard deviation; use s if unknown in the large-sample approximation
- Random sample size
Parametric VaR = one-sided Z-value × return standard deviation × position value
- One-sided critical Z-value
- Return standard deviation over the VaR horizon
- Current position value
For example, a random sample of 64 loans has mean balance EUR 64,000 and sample standard deviation EUR 52,000. Using the lesson’s large-sample normal approximation, SE = 52,000 / √64 = EUR 6,500 and the 95% interval is 64,000 ± 1.96 × 6,500, or EUR 51,260 to EUR 76,740.
Common mistakes
- Using individual-loan standard deviation directly in the confidence-interval width is incorrect.
- Dropping the critical value leaves a range that is not a 95% confidence interval.
- Using a two-sided critical value for one-sided VaR overstates the lesson’s VaR.
Try a practice question
Choose an answer, then check it to read the explanation for every option.
Continue studying in Quizara
In Quizara the lesson comes with more practice questions on the topic, explanations for every option and an AI tutor for your follow-up questions.
Written by the Quizara Editorial Team under our editorial standards.