This code computes the LR test statistics proposed by Christoffersen (1998) in the context of the backtesting of Value-at-Risk forecasts. The tests are then based on the concept of violation: a violation is said to occur when the ex-post losses are larger than the VaR defined for a given coverage rate. The user can choose to test for Unconditional Coverage (UC), Independence (IND) and/or Conditional Coverage (CC) assumptions. The two first corresponding test statistics have a chi-squared distribution with one degree of freedom and the last one a chi-squared distribution with two degrees of freedom. For more details, please read the following document.
Perignon Christophe and Hurlin Christophe