A lower significance level ⇒ greater confidence interval ⇒ higher absolute VAR For VAR we only care about the lower tail of the distribution. 5% VAR ⇒ 90% Confidence Interval ⇒ 1.65 standard deviations below expected return 2.5% VAR ⇒ 95% Confidence Interval ⇒ 1.96 standard deviations below expected return .5% VAR ⇒ 99% Confidence Interval ⇒ 2.58 standard deviations below expected return […]

deviations of a portfolio’s factor sensitivities from the benchmark factor sensitivities.

The long position is exposed to potential credit risk in a payer swaption at initiation, but the short position in a payer swaption is not.

Risk budgeting occurs when the firm chooses a desired maximum risk and then budgets this out to individual portfolios. The risk will generally be based on ex ante tracking error or VaR.

A VAR statistic had 3 components: a time period a confidence interval a loss amount (or loss percentage) There are 3 ways to calculate it: Historical Method using a histogram Variance-Covariance Method assumes normal distribution Monte Carlo Simulation Worst case is worse than VAR confidence interval predicts.

TRUE It can be diversified away.