When we specify something like a 95% value at risk or 95% VaR, we mean that 95% is the confidence level and, therefore, 5% is the significance level. That means we expect on 5% of days for the actual loss to be worse than the VaR or to exceed the VaR. This video is about the backtest of a VaR, which is a very handy statistical tool that we have at our disposal to backtest or to test the validity of this model. The backtest will allow us to make a decision as to whether to accept this as a good VaR model or to reject it as a var model.
