Reference book for interest rates modeling?

Qwer Ty

New Member
Hi,

I'm interviewing for a rates modeling quant role in a sell side bank. The role is centered around pricing and risk management of rates trading carried out by the front office.

I've been told to prepare for the technical aspect of the interview, with emphasis on topics like risk-free pricing, risk-neutral pricing, swaptions, etc. I've also been told to refer to books like Andersen and Piterbarg, Brigo Mercurio to learn these topics from.

I don't have any background in finance, and I have an undergrad degree in engineering (think Chem/Mech/Civil). The problem I'm facing is, texts like Andersen, Brigo are coming across as advanced or high level, and I can easily feel some knowledge gap between my own level and whatever these texts contain. I started refering to some lecture notes by Damir Filipovic who takes a coursera course on Interest Rates Modeling, but even his notes are on some intermediate/advanced level.

I'm searching for some reference book that starts from fundamentals that explains basic concepts and only then becomes advanced, helps me build an intuition, and gets into interest rates modeling from thereon.
 
Baxter & Rennie is I think the best intro book; it's not too rigorous (I find the rigor of Shreve to be more distracting than useful in an intro book, though I'm sure many here would disagree with me) and it is short (unlike Hull, which tries to cover way more products-wise). After B&R you should be able to approach A&P (though if you don't have anyone to tell you which parts you should read first, it's still a bit heavy as it's a bit of a reference book).
 

Qwer Ty

New Member
I just realized that a lot of these topics are covered in the CFA L2 L3. Does anyone think that using the CFA books to study these topics would be a good idea, rather than using these reference books?
 

rajanS

Active Member
U need to read that book by lixin wu and another by Nicholas privault. Both are excellent for breakin into the Ir world. I don’t understand why this bank is even calling you when you don’t have knowledge in this area.
 

IntoDarkness

Active Member
U need to read that book by lixin wu and another by Nicholas privault. Both are excellent for breakin into the Ir world. I don’t understand why this bank is even calling you when you don’t have knowledge in this area.
this... these two books are usually enough for most jobs
 

pingu

Well-Known Member

Razvan Ilie

Member
C++ Student
I'm reading Brigo and Mercurio right now. How much overlap is there between books? Say with other ones mentioned here like Andersen & Piterbarg, Wu, Privault, etc. Are they mostly different perspectives on the same material or do they present mostly different material?
 
Last edited:

rajanS

Active Member
I'm reading Brigo and Mercurio right now. How much overlap is there between books? Say with other ones mentioned here like Andersen & Piterbarg, Wu, Privault, etc. Are they mostly different perspectives on the same material or do they present mostly different material?
if you understand those books...ur fine. U should be aiming at implementing the models in these books. Not just reading them.
 

Daniel Duffy

C++ author, trainer
I'm reading Brigo and Mercurio right now. How much overlap is there between books? Say with other ones mentioned here like Andersen & Piterbarg, Wu, Privault, etc. Are they mostly different perspectives on the same material or do they present mostly different material?
a nice C++ implementation using decomposition methods (like my Mont Carlo design). Just an idea.
 

Razvan Ilie

Member
C++ Student
a nice C++ implementation using decomposition methods (like my Mont Carlo design). Just an idea.
I was actually just thinking that. The final project from the advanced course pretty much allows ultra fast implementation of any one factor model. Perhaps the next step is upgrading the framework to accept two factor models and also calculate implied volatility with confidence interval and maybe risks as well
 

Daniel Duffy

C++ author, trainer
I was actually just thinking that. The final project from the advanced course pretty much allows ultra fast implementation of any one factor model. Perhaps the next step is upgrading the framework to accept two factor models and also calculate implied volatility with confidence interval and maybe risks as well
Super idea. The inter-component function signatures will be different but the overall structure will remain the same. You can always contact me if you have questions. Good luck :) It would be a nice achievement as well as a reference design for many other 2-factior problems.

I would avoid too much inheritance (CRTP is good) and use black box interfaces std::function. The data flow all the way from input to output will get you there.

Try not to embed hard-coded 2-factor knowledge in classes. Avoid using tuples to create/hold/pack/unpack run-time data, it's shlow.
 

Attachments

Last edited:
Top