• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

The money formula

vertigo

vertigo

Joined
3/19/14
Messages
95
Points
43
New book by paul wilmott called the money formula released recently. Has anyone read it? I have read about half of it and it is crap. Not informative, not useful and the more I read of it the more I realise why nobody takes Wilmott seriously - his views are very naive and, surprisingly, stupid. It's like the guy has never worked for an investment bank.
 
Thank you for providing the link to the book, Andy. Here is a quick summary of what the book gets wrong in my view. Keep in mind I have not read all of the book, but I have read more than half of it.
  • The authors advise using modelling tools from mathematical biology to solve problems in finance. For some reason and I don’t know why, the authors think that using mathematical biology models will help to solve existing problems in finance. They do identify problems that current models do not pick up, such as the memory and time lag (for example when breaking news is released on TV, naturally there will be a time lag for markets to pick up on it). They advise reading an old book on mathematical models for biology, saying it is the best book ever for mathematical models… Just bizarre. You cannot take an author seriously, especially in the world of banking, when they say things like that. On the topic of memory, a way to capture memory is by using a subordinated time changed Levy process , i.e., instead of time t you use a modified time t_1 = g(t). That is not a mathematical biology model, it is a mathematical finance model.
  • They ramble about the failings of the efficient market hypothesis and how quants are mesmerised by the EMH. But when Wilmott runs a survey on how many quants believe the EMH, it is only 43%, which is a good number, I think. How can they be mesmerised when less than half believe it? They are not happy with that number, they want it to be zero. It is a stupid position to take because people who criticised non EMH methods (e.g., behavioural economics, which is also crap) also want the percentage of peopel who do not believe in EMH methods to be zero. Hey, this is reminiscient of maximising your expected utility and not caring about what others think, which goes against behavioural economics, but let's forget that.
  • They are baffled that more actuaries are learning about quant methods as opposed to the quants learning about actuarial methods. This is not just stupid, it is wrong, I have taken the actuarial exams and some of the exams, namely business economics and financial economics – teach the methods they despise of so much, i.e., they promote the tools that are used in EMH, CAPM, APT and FTAP (e.g., Black Scholes). In fact one of the best things a young quant can do is finish some exam papers in the module CT8 Financial Economics that are freely available from the Actuarial Society – they have lots of practical, relevant and interesting questions about using toy models for financial derivatives. But this goes against what they are saying. They think that quants should learn things like linear regression, contingency tables, etc… if you advise using linear regression models for derivative pricing, you need to be fired straight away.
  • They advise using methods from the P-world (real world) because, in my view, they do not understand Q-world (risk neutral). More specifically, they believe the asset price processes should perfectly match reality – skew, non-continuity, kurtosis and so on… This shows they do not understand hedging post 1987 because the hedges are not the underlying, but the actual vanillas are. Vanilla prices are modelled well by Black Scholes – they mention a few lines that hedging for vanillas is actually accurate – page 143 - but beg for more information on this. They should get their eyes out of their asses and start looking because many quants have written papers on what are the correct hedge ratios to use. They also think models should ‘predict’ the markets, i.e., model prices should forecast future market prices. But if you ask any quant or trader, they would tell you that they do not use models to predict market prices. They use models to generate sensible hedge ratios. Neither author has worked at an investment bank (or even had a boss, in my view, because they are living in their own world, quants are far more practical and pragmatic than either author) because the view that pricing models should predict market prices only comes from people who haven't worked in investment banks.
  • They claim that quants do not know about randomness or risk neutrality, for example they think most quants believe in complete markets, but actually many books on using non-Gaussian processes to model derivatives (e.g., Levy processes, jump-diffusion), i.e., models for non-complete markets, both on the theoretical and practical level, that it shows they are dogmatic about their view despite evidence existing that goes against their view.
  • They think there are no fundamental laws in economics. I don’t even know what to say about that. It is stupid, it is wrong and it is reminiscent of somebody who just hates the concept of mathematics being used in economics. Of course there are fundamental laws in economics, for example the law of one price, exchange rates not being negative, etc….
  • They criticise mathematical finance master programs as being rebranded versions of measure theory. This is also incorrect and people on this forum can probably verify that. Mfin master programs get (at least one or two) practitioners from the industry to teach modules or at least provide one or two lectures, which goes against their narrative that Mfin master programs are heavy on mathematics, light on practical applications. But this is a contradicting position to take because the models they advocate for are even heavier on mathematics and even lighter on practical applications!

This book is not marketed to experienced quants or even experienced investment bankers… This is not a major problem because the poison they spout is not taken seriously by experienced quants, but young quants, perhaps coming from university or currently at university and wanting to read up on how the quant world and markets work, are more likely to be affected by this poison and end up hating the quant industry. I really dislike this view, because working as a quant in an investment bank is fantastic for young scientists. Even if the industry is not perfect for you, there are many things you will learn - how to work in a fast pace work environment, work with a diverse set of people (traders, sales, project manangers, etc..), these are things that are applicable in most industries.

P.S. Wilmott offers a certificate in quantitative finance that promotes the use of models he so despises – having seen the CQF and the models they use, they are actually far behind most investment banks, which goes against his narrative that banks use crap models. They don't mention this, of course.
 
Back
Top