I used to do this project in university and our method do not incorporate the ois we did extrapolation instead.

Won't make too much difference if dt is very small.Implementations for both are not tedious anyway.You can referred mostly to Jeff Greco's Notes.Hope this is helpful.

<t>QuoteOriginally posted by: mathmarcQuoteOriginally posted by: JohnGuThat's the method I am using.But the value i get is quite off.I used the gradient function in matlab.Why do you want to compute instantaneous forwards? They are a useful theoretical concept, but I have never seen a practical case...

That's the method I am using.But the value i get is quite off.I used the gradient function in matlab.

Hi all, a question is to calculate the instantaneous forward rate of future time t , e.g f(0,t).Data we have are the zero rates of different maturities.I tried the gradient function in matlab but the results were quite off.Any help is appreciated! Thank you.John.

Thx for the reply. I found they are also using spline interpolation and then to find derivative.But this method would gave back very off results.And I am looking for some better methods to calculate a(t) that use analytical formula instead of using forward induction.Thank you.

<t>Let me try to answer it if I understand your question correctly.For CIR, you have three parameters, and the current yield curve is to calibrate the parameter in the drift term.you could have another parameter after one year. The CIR model is not for forecasting, but pricing.And the expected short...

<t>Hi all,I have a question on the Hull White tree.They used a forward induction for the tree but in Brigo's book, he mentioned the analytical formula could also work.But how could we get the instantaneous forward rate from the current yield curve?My method is to use the matlab function spline to in...

Clewlow and Strickland is a good place to start as their matlab code is available online.But do bear in mind that there's implementation error for their hull white.

Duffie and Kan(1996) called those uninterpretted variables latent variables.So it depends on how to what's your degree of freedom.

I tried the code from volopta,But the fminsearch would gave me different values for alpha and sigma if use different starting values.Don't understand why.

hmm good idea, but I don't have the implementation for time varying alpha or sigma.Do you know where can I find it?Thank you very much.John.

Hi, I have build the tree given the yield with different maturities.So how could I modify my tree to match the cap price if I want to calibrate \alpha(t)? Thank you.

<t>Hi all,I have a question on the general one factor hull white with all parameters are piecewise constant time dependent.[$]dr = (\theta(t)-\alpha(t))*dt + \sigma(t)*dW[$]So [$] \theta(t) [$] is to fit the initial yield curve, but how could we decide the other two parameters (function)?Say one swa...

Thank you all, but one more question.In Brigo's book, he said for the second stage for the trinomial tree building, we add a alpha to obtain the tree for r. If we use analytical formula for alpha,we could not get the inital market yield curve.I don't know why is this so?