A Simple General Approach to Inference About the Tail of a Distribution
THE TAIL OF A DISTRIBUTION 1169 yr+1) > D, the conditional likelihood function for 0 is (3.3) L1(6) oc IJlexp[k In w(y(l'; 0) - (k - i) In [w(y(i); O)/w(y(i+?'; 0)]] where the Jacobian J is proportional to Jl+1 (dIn w(y'i'; 0))/dy'i'. It is interesting to...