In the standard LMM a rate $L_i(t)=L(t,T_{i-1},T_i)$ has under the $T_n$-forward measure ($n>i$) the dynamics \begin{equation} d{L_i}(t) = - {\sigma _i}(t){L_i}(t)\sum\limits_{j = i + 1}^n {\frac{{{\tau _j}{L_j}(t)}}{{1 + {\tau _j}{L_j}(t)}}{\sigma _j}(t){\rho _{ij}}} dt + {\sigma _i}(t){L_i}(t)d{W_i}(t). \end{equation} Now that the rates are negative, it would be desired to model a displaced diffusion, i.e. instead of $dL_i$ model $d(L_i+\gamma)$.

Is such an adjustment possible without breaching the model arbitrage-free assumptions?

I already checked the literature and nowhere I can find any references.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.