Imagine that one day, you see people are queuing up in front of Bank A; so you ask the staff at the counter, you are told that they are offering anyone (regardless of their credit history) a loan of $100,000 at a fixed annual rate at 2%. You then look around, the Bank B next door offers 1-year term deposit with a fixed annual rate at 3% for the same amount ($100,000). After 5 minutes' waiting, you sign for the loan from Bank A and immediately, you deposit the borrowed $100,000 in Bank B, expecting to earn 1,000 dollar based on (3%-2%) x 100,000 one year from now. Your lucky day!
In Finance, without investments, an opportunity that allows one to gain profits from simultaneously buying and selling something is called Arbitrage Opportunity. The above opportunity, which involves borrowing through a loan from Bank A and lending through a deposit at Bank B, represents an arbitrage opportunity in financial economists' parlance. However, one can easily see this type of opportunities won't last long. In efficient markets, according to Nobel Laureate economist Eugene Fama, such arbitrage opportunity should not exist and could never last. However, in our paper titled Arbitrage opportunities in CDS Term Structure: Theory and Implications on OTC Derivatives, we found that the contrary is true at least in the credit default swap market (to be explained below).
In this study, we applied calculus, probability theory and economic principles to derive so-called No-arbitrage conditions, which basically sets out the basic relationship that should be followed by CDS contracts of different maturities should follow. Violating the conditions we derived, nonsensical probability (negative or greater-than-1 probability) and arbitrage opportunities will emerge. Based on the mathematical conditions derived in our research paper, after examining 30+ million CDS contracts, we identified 2,416 pairs of CDS contracts as arbitrage opportunities, which are not supposed to exist. Is the Nobel Laureate wrong or are there other reasons that deter investors from queuing up to take advantage of such opportunities?
CDS is basically tradable insurance contract that provides the buyer the recovery of pre-specified losses as the result of the reference name. It is slightly more complicated than the loan in the example above; however, the principle of arbitrage works exactly the same. Here is an example from two CDS contracts selling CDS protection against default of Microsoft (Please read the paper for more details.). The green line indicates the value of the trades. On 03/12/2008 (Please note that the notorious Lehman Brothers just defaulted 3 months earlier on 15/09/2008; it's the prime of the financial crisis.), the green line starts at zero, indicating that no investment is needed; also, throughout the time, it is above zero, which means positive profits persist for the trades, thus, an arbitrage opportunity. Further details are in the paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3282399
Views: 1687
Tags: Analytics, Big, Calculus, Data, Finance, Mathematics, Science, Statistics, Stochastic
Comment
Hi Vincent,
Fantastic book and thanks for sharing! Real-world finance problems depend heavily on understanding and applying stochastic processes, your book is a great addition to the literature from someone with a lot of real-world experience. Purists like to think the stock market as well as other financial markets are so-called Markovian, i.e., future prices are only dependent on today's information set, not anything further back, but actually they are and they are not. That's the distinction between Efficient Market hypothesis advocated by Prof. Eugene Fama and those by Robert Shiller. Both of them won Nobel prizes in the same year.
Without wading into their lifelong debate, my research focuses on one segment of financial market, CDS market; as you know, a lot of banks which traded with it had problems during the 2007-09 crisis; the CDS market, as evidenced by our research, is far from Absence of Arbitrage (AoA), which is the core assumption underpinning the practices and theory in today's banking industry and finance being taught at universities. However, throwing away AoA would make valuation difficult, which is related to the uniqueness problem (like in math); i.e., you won't have unique value for financial derivatives any more, more theoretically, the unique risk-neutral probability measure.
Congrats on your book and best regards,
Zhongmin
p.s.: anyone interested, please go to my home page and download my papers originated from my academic research; if helpful, please feel free to cite my papers. many thanks.
https://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=2064225
Hi Zhongmin,
There is one thing in this chart that surprises me a lot. I have produced plenty of charts that look very similar to yours, but come from my number theory investigations. I am sure some of them can be found in my new book on stochastic processes. I came to the conclusion that there are many things in number theory (advanced probabilistic properties of numeration systems or prime numbers, in particular) that are very valuable for Fintech modeling. Indeed, my book, which is a book about number theory (in disguise), mentions many applications to stock market modeling, Brownian motions, and better stochastic processes (derived from the theory of numbers) to explain stock price fluctuations. For instance, the digits of some numbers (say Pi) behave in a such a way that they appear totally unpredictable -- indeed they can be used to build Brownian motions -- it just looks like those digits have the same amount of "arbitraging" (or call it entropy) in their construction, as the stock market. Yet, you can derive significant value from them. Afterall, they are not random at all even if they appear to be so. Same with the stock market!
Vincent
© 2019 Data Science Central ® Powered by
Badges | Report an Issue | Privacy Policy | Terms of Service
Most Popular Content on DSC
To not miss this type of content in the future, subscribe to our newsletter.
Other popular resources
Archives: 2008-2014 | 2015-2016 | 2017-2019 | Book 1 | Book 2 | More
Most popular articles
You need to be a member of Data Science Central to add comments!
Join Data Science Central