Poll: for how much would you buy a book on "Getting Started with QuantLib"?

[yetanotherquant]

Hello together,

I am thinking about writing an introductory book on QuantLib.
I am going to assume only the basic knowledge of financial math and C++ (or even C, without OOP).

I am not a core developer of QuantLib but I have a long experience with it.
In particular, in this thread I am the poster on NP, to whom the credit goes.

Though I do not expect to earn much with this book, I am not going to do it "just for glory".
So I would like to study the market first

 

[Daniel Duffy]

Hello together,

I am thinking about writing an introductory book on QuantLib.
I am going to assume only the basic knowledge of financial math and C++ (or even C, without OOP).

I am not a core developer of QuantLib but I have a long experience with it.
In particular, in this thread I am the poster on NP, to whom the credit goes.

Though I do not expect to earn much with this book, I am not going to do it "just for glory".
So I would like to study the market first

Will you be in Dusseldorf 4-5 December 2014?

http://quantlib.org/qlws14.shtml

I'll be there

BTW how's your C++ fluency?

 

[ExSan]

PDF books should cost not more than $3.00,
it would be the only way I would make a reality my Amazon wish list

 

[yetanotherquant]

Will you be in Dusseldorf 4-5 December 2014?

Unfortunately no
I would really like but our teamleader is on vacation and I have to deputize for him.

BTW how's your C++ fluency?

Achtung! This is very interesting question.
Short answer in context of this poll: good enough to solve my tasks by means of QuantLib as well to extend it in such a way that it works for me (but not necessarily always good enough to be contributed to QuantLib repository).

A broader answer:
1) no one is perfectly fluent in C++, even Stroustrup is not!
1a) In particular, if you have an interview, be ready to get a question you cannot (correctly) answer.
However, you can put a counter question to your interviewer (whether it increases your chance to get a job is another issue).
My favourite one ist to ask what modifier "const" means. So far noone was able to name all 3 meanings.
2) In order to solve your tasks as a quant, you do not necessarily need to be a C++ guru. Moreover, C++ as such is only a part of iceberg. E.g. with QuantLib one problem, which one encounters very soon, is to represent the objects of the Date class in a readable form in Debugger (for those who understand it: autoexp.dat was not really helpful; I, myself, do not know yet whether it can be solved my means of new VS2013/13 features like natvis)

 

[yetanotherquant]

PDF books should cost not more than $3.00,

Let us count a little bit
An hour of a professional like me costs at least $100. Since writing/teaching is my hobby, let's set it to $50.
Experience shows, I need ca 1 hour per page. Ok, the book in question will likely contain a lot of screenshots, so let's say 2 pages per hour. The book will be (at least) 200 pages, so the total cost is 200/2 * $50 = $5000.
So I need to sell 5000/3 = 1666 copies just in order to be breakeven. Moreover, Amazon also wants its share, so what I get will likely be not more than $1 per book. Thus 5000 copies would be needed...

 

[zyz]

pro bono it and make donation available

 

[yetanotherquant]

pro bono it and make donation available

Unfortunately, my experience speaks against it.
I wrote some good stuff, in particular the LIBOR Market Model Tutorial (s. attachment).
I didn't even ask for donation but I expected people will at least say "thank you".
Very few did, though I get several downloads every day.

BTW, Luigi himself reports similar problem: "I see thousands of people subscribed to the mailing list, with many subscribers from big name institutions, and I see thousands of downloads every time we put out a new release but people doesn't tell, and there's just not that much data around on actual practical applications, or whether people use it in production"

I think people just undervalue free stuff. On the other hand, if they pay money for a book (even a little bit), they usually read it - because they need to justify this money, so to say.

 

[Daniel Duffy]

What kind of books would you to write (cookbook?) and for whom?

 

[yetanotherquant]

>for whom?
First of all, for MFE graduates and advanced undergraduates.
They are most numerous and some of them are (still) curious for knowledge.

>cookbook?
Yes, mostly (what else can one write for graduates?!)
However, I will also try to show the way for a cook, who wants to become a food-biochemist

 

[Daniel Duffy]

>cookbook?
Yes, mostly (what else can one write for graduates?!)

The answer would come from a market research.

 

[IntoDarkness]

if its a really good one (somewhat rigorous theory + various illustrated example + several big projects that integrate lots of parts in the book) i wouldnt mind paying $30.

 

[Daniel Duffy]

if its a really good one (somewhat rigorous theory + various illustrated example + several big projects that integrate lots of parts in the book) i wouldnt mind paying $30.

+- 90 pages

 

[pingu]

Are you going to self publish?

 

[yetanotherquant]

Are you going to self publish?

Most likely, yes

 

[yetanotherquant]

Following a hint on Wilmott here is a very approximate book plan:
0) How to build Boost + QuantLib
1) (optional) brief review of OOP fundamentals
2) Boost-based components of QuantLib, in particular smart pointers. How to visualize them in VS-debugger.Design patterns (Dimitri Reiswich has already done an excellent job, but still...). UnitTests. Creating and reading Doxygen-Doku.
3) Interest rate modeling from very beginning and very concretely (as Quant I am mostly experienced with fixed income) : Date arithmetic (daycounters, business day conventions, etc).Bonds. Fitting yield curves: NSS, Cubic Spilnes, etc.
3a) Interest rate models (equilibrium vs. no arbitrage, short rate vs. Libor, calibrated models, affine models, term structure consistent models, etc)
4) Relation of financial instruments, pricing engines and models in QuantLib
5) Pricing [simple] instruments
6) [Discrete] Hedging revisited (specially for graduates: why it is important to hedge, not just to calculate the price as risk-neutral expectation of the discounted payoff).
7) (optional) some modern/advanced topics: CVA/DVA adjustment, multicurve framework, etc. (optional) integrating QuantLib and CUDA/GPU for ad-hoc tasks.

 

[yetanotherquant]

By the way, as I was a recent graduate and looked for job, I got a homework after a phone interview (s. attachment). The 1st exercise was a piece of cake for me (due to "round" maturities). But the second one was the barrier I did not pass ... I do not mean that I failed to solve it. I offered some plausible approaches (that were, however, non-standard oppositely to NSS or CubicSplines). I am still in touch with the head of dept, who made that interview and he actually wanted a kind of linear interpolation from me.

But I think if I had applied all interpolation methods available in QuantLib and told him "choose whatever you like", I would probably get the job.
Alas, there was noone who wrote "Getting started with QuantLib" by that time :P

 

[Daniel Duffy]

I am still in touch with the head of dept, who made that interview and he actually wanted a kind of linear interpolation from me.

Arbitrage?

 

[ExSan]

Following a hint on Wilmott here is a very approximate book plan:
0) How to build Boost + QuantLib
1) (optional) brief review of OOP fundamentals
2) Boost-based components of QuantLib, in particular smart pointers. How to visualize them in VS-debugger.Design patterns (Dimitri Reiswich has already done an excellent job, but still...). UnitTests. Creating and reading Doxygen-Doku.
3) Interest rate modeling from very beginning and very concretely (as Quant I am mostly experienced with fixed income) : Date arithmetic (daycounters, business day conventions, etc).Bonds. Fitting yield curves: NSS, Cubic Spilnes, etc.
3a) Interest rate models (equilibrium vs. no arbitrage, short rate vs. Libor, calibrated models, affine models, term structure consistent models, etc)
4) Relation of financial instruments, pricing engines and models in QuantLib
5) Pricing [simple] instruments
6) [Discrete] Hedging revisited (specially for graduates: why it is important to hedge, not just to calculate the price as risk-neutral expectation of the discounted payoff).
7) (optional) some modern/advanced topics: CVA/DVA adjustment, multicurve framework, etc. (optional) integrating QuantLib and CUDA/GPU for ad-hoc tasks.

I am interested in your book and wouldn't mind the price, I am particular interested in 7) and BTW would you include HF trade/algos ?
salut
+XSN

 

[yetanotherquant]

Arbitrage?

Very likely since this bank indeed went bankrupt a couple of years ago
Ok, seriously, in theory it is relatively easy to construct arbirtrage from a (mis)interpolated yield curve but in practice (bid-ask spreads, limited ability of bonds) it is much harder ...
On the other hand I witnessed how some guys made money on the imperfect fit of the NSS approach
(it was not arbitrage but rather StatArb, which worked well).

 

[yetanotherquant]

I am interested in your book and wouldn't mind the price, I am particular interested in 7) and BTW would you include HF trade/algos ?
salut
+XSN

Thanks for factual input and encouraging comment.
As to HFT algos, unfortunately I never did it, so I would rather avoid a danger of being Irene Albridge II
And as far as I know from papers and from the guys who did, CUDA/GPU find limited application to HFT.
FPGA are much more common but I never programmed them, at least so far.