Description:
Quantitative Finance Collector is a blog on Quantitative finance analysis, methods in mathematical finance focusing on derivative pricing, quantitative trading and quantitative risk management.
Contents:
Kalman Filter Week in Review 090212
A tractable LIBOR model with default risk:a model for the dynamic evolution of default-free and defaultable interest rates in a LIBOR framework.
Optimising a correlated asset calculation on MATLAB:detailed example of applying vectorisation to speed up Matlab codes.
Reading About the Financial Crisis: A 21-Book Review: Professor Andrew W. Lo reviews a diverse set of 21 books on the crisis, 11 written by academics, and 10 written by journalists and one former Treasury Secretary. Are they helpful to understand the current crisis?
A Forward Monte Carlo Method for American Options Pricing: This study proposes a forward Monte Carlo method for the pricing of American options, and significantly improves in numerical efficiency and accuracy in contrast with the standard regression-based method of Longstaff and Schwartz(2001).
ReBEL : Recursive Bayesian Estimation Library and Toolkit for Matlab: I couldn't find a good R package for extended Kalman Filter parameter estimation, ReBEL is one for Matlab though. Please let me know if you know some R package. Three excellent papers to understand Kalman Filter in finance are:
1. Estimating and Testing Exponential-Affine Term Structure Models by Kalman Filter, 1999, Jin-Chuan Duan and Jean-Guy Simonato, REVIEW OF QUANTITATIVE FINANCE AND ACCOUNTING
2. Affine Term-Structure Models:Theory and Implementation, 2001, Bolder, David Jamieson, Bank of Canada Working Paper No. 2001-15
3. Non-Linear Kalman Filtering Techniques for Term-Structure Models, 1997, Jesper Lund, Working Paper
Exploiting Option Information in the Equity Market: Strategies based on several option measures predict returns and alphas on the underlying stock.
Interview: Thijs Van Den Berg From Sitmo.com: a manager of Sitmo B.V founded in 1998.
Tags - libor , default , matlab , crisis , monte carlo , american , kalman-filter , option
Unclear about this post? Read the full post at Kalman Filter Week in Review 090212 or Asking questions and receiving
answers.
---supported by Best selling investing books 
Interview: Thijs Van Den Berg From Sitmo.com
Standing on the shoulders of giants allows us to see further, from now on we will invite experts to share with us their valuable experience and lessons.
It is our great pleasure to have Thijs van den Berg joining this week's interview session, Thijs is the manager of Sitmo B.V founded in 1998, which was initially a derivative market-making firm operating on the European Options Exchange (now Euronext), but soon building customized derivative models and risk management software development became an important activity. In 2003 Sitmo started consultancy services in Energy trading and quantitative modeling.
Tell us a little background info about yourself. Where are you from? Whats your education background?
Im from The Netherlands. As long as I can remember Ive been curious: math, physics. I got my first computer when I was 10 and things became magical: I had my personal desktop lab to experiment with! About that time my family decided to move to a sunny island. I had a great time windsurfing, surfing and skating, but education was a bit 2nd place. I went a year to a local Spanish school but didnt speak much Spanish and so the only thing I could follow was the math classes. The second year I went to a British International school and that was very intense and good. Every morning sausages and beans etc. After that we moved back to The Netherlands, I skipped a school year, and eventually went to the Delft Technical University when I was 17 to study Computer Science. The first year was perfect -I was in the top 5-, but then I started to doubt my choices... I ended up working in a popular bar and was really enjoying that, ..until a professor knocked on my door and said he wanted to talk to me. Hes now a very good friend. After that I quickly finished university, did a thesis at a bank on forecasting with Wavelets.
Do you have any experience with quantitative finance? If yes, how long have you been in the quantitative finance industry and to what extent?
I ran into QF when I started trading (equity) derivates on the floor in the 90s. Id build our own option pricing models and risk management tools, those were great times, we always had different prices than other traders, but we got it right... After that I got a job running a quant department at an energy Company. Energy trading was in its infancy: there was extremely much to do from a modeling perspective. The commodities have very complex dynamics, exotic assets, optimization, load forecasting, credit, data warehouses. We managed to get a couple of good PhD on board who delivered good models on fundamental activities. It was a true startup: when I joined the company the trade floor was just 6 people, when I left 250 with full blows specialized departments.
I left there in 2003 and I started my consultancy firm and have been doing that ever since. I took one year off in 2008 setting up an algo trading firm with two partners. It was very heavy on computing, scalability, reliability, data mining, stream processing, exactly replicating exchange matching engines etc,.. lots and lots of C++. It was great fun, very long days, but we made a fatal mistake at the beginning with the contract. It ended up being a big write-off, and an expensive lesson learned.
What is your specialty? Equity, fixed income, derivatives or others, and to what extent?
I think my specialty is more my drive to perform and solve. I enjoy learning new things very much, and I seek new types of problems whenever I can. In general I do a lot of strategic advice, energy, coding and trading related projects. My last project was completely different: a 7 month fixed income model validation and liquidity modeling task.
What accomplishments so far are you the most proud of?
Ive build a modeling framework that can learn complex non-linear dynamic from observation data. It performs extremely well and it has a very elegant mathematical foundation. Its very versatile and I use it to model complex time-series like energy spot stochastic, volatility term structures, intra-day FX, even weather dynamics like temperature, wind speed and light intensity. I think Ive invested at least 4 years in developing the math and coding.
Why did you choose this career? What are the pros and cons of working as an independent consultant instead of in a big company?
My parent had their own firm, and thats one of the first reasons. Another reason is probably my personality: I enjoy initiating things, I thrive on new knowledge and solving complex issues, and Im not risk adverse, I like to challenge things and innovate. The cons are the unclear distinction between work and home. Clients are expecting a lot of performance, and there are always tight deadlines. Socially its of-course also a bit different, you come and go. My colleagues are probably my peer consultants I regularly team up with on projects.
What do you think it takes to be successful as a quantitative analyst?
Be honest to yourself. Never bend figures towards a predefined goal by anyone. When modeling: know about model error and over fitting, always try to validate results with common sense back-of-the-matchbox bounds and simple proxy models. Know the difference between accuracy and precision. Another aspect of being honest to yourself: when youre wrong directly say that you were wrong, it takes guts to do that. People will respect that, and it allows you to move forward faster. Time management is also very important. Continuously try to deliver small increment, dont hide in a closet for half a year.
What is the single toughest challenge youve had to face in your past projects, and how did you get through it?
The toughest challenge was working in an extremely political war-type of environment and try to be productive. Almost all divisions at my client were in serious conflicts with each other, managers trying to get each other fired, lying, put the blame of failures on each other, there were coupes. No one was looking after the company. I had to pick a couple of battles and those were mainly on establishing clear boundaries professional and making clear that I dont accept certain type of behavior.
What is the future of quantitative finance in your opinion, especially after the financial crisis?
My opinion is that there will always be need for improving things. Saving money for a company, helping make better decisions, help them value things more precisely, help reduce risk and save on capital needs. These activities will always be valuable.
What have you been up to recently? What projects are you working on?
Many things in 2011. For a bank Ive done model validation, setting up swap curves and building liquidity management models. Ive done a strategic advice for a large international utility that wanted to quantify their strategic lobby possibilities. Ive started an open-source Quantitative Finance Code Library Platform with Paul Wilmott and Daniel Duffy. Ive build optimization and risk models for a large group of CHP owners to help them optimize their operation and manage their risk. Another optimization modeling project was for the water utility. They have storage buffers, strict safety bounds and use lots of energy to process water and pump it around the country.
What is the best advice youve been given and you like to share with Quant wannabe?
During a review, ask your managers about the things you are not good at and then dont try to get upset about that but instead think and talk about it. These things typically transcendent the workplace and not many people tell you about your flaws in your life.
How do you like to spend your free time?
My family is vey important to me, I like to not just work 24/7 but also be at home, I value that highly. I run to keep fit, I still have a skateboard and there is a skate park around the corner. I enjoy playing pool with my friend, go with then to camp at festivals and see bands. Another thing I like it those relaxing spas with my wife -but I hardly do that-.
Do you have other suggestions you like to share with us?
Try to be happy and have a diversified identity, you only live once.
How can people contact you for consulting business? Do you have a website or Twitter account or Facebook Like page?
Website: www.sitmo.com, email: thijs@sitmo.com, twitter: sitmo_com
Tags - interview , consultant , quant
Unclear about this post? Read the full post at Interview: Thijs Van Den Berg From Sitmo.com or Asking questions and receiving
answers.
---supported by Best selling investing books
QuantShare Trading Software
The manager of QuantShare has contacted me about this software, after several days trial I feel it may be of interest to some of you so I post a short introduction here.
QuantShare is a new technical/fundamental analysis software available since only few months.
The sharing server is what makes QuantShare apart from anything else. It is a place where users can share their trading systems, indicators, downloaders, custom drawing tools...
If you need intraday data for futures, simply search for a downloader in the sharing server and chances are you will find one already implement by a member of the community. There are more than 800 items there and this number keeps increasing every day.
Besides, the sharing server, QuantShare has an impressive number of tools (Charting, Simulator, Composite, Genetic Algorithm, Neural Network ...).
The simulator for example allows you to create trading systems and backtest them very easily. The money management tool can be used in case you want to implement strategies that are more advanced. Once your system is ready, the Portfolio tool allows you to generate buy and sell orders automatically using the trading system rules.
It is also possible to include composites, fundamental data, news data and anything that can be quantified in your system.
Click here to get a free QuantShare Trading Software trial.
Tags - trading , software
Unclear about this post? Read the full post at QuantShare Trading Software or Asking questions and receiving
answers.
---supported by Best selling investing books
Week in Review 020212 Quantitative Finance
A Sea Change in Quantitative Finance: thoughts on P - Q Convergence in Quantitative Finance.
An Alternative Three-Factor Model: A new factor model consisting of the market factor, an investment factor, and a return-on-equity factor reduces the magnitude of the abnormal returns of a wide range of anomalies-based trading strategies.
People of Quant Research: a list of influential people in academy on Quantitative Finance research.
What Strategy Worked in 2011: what might cause the different performance of funds in 2011, is it due to trading strategies?
Bloomberg Open Market Data: Now you can adopt Bloomberg's market data interfaces without cost or restriction.
Kalman Filtering in R: Pros and Cons of existing R packages for Kalman Filtering.
Volatility-responsive asset allocation: a volatility-responsive asset allocation policy can lead to a more consistent outcome and a better trade-off between risk and return.
China, Merkel and Euro: Merkel knocks at China's door, with poor Euro.
Tags - quant , factor , strategy , bloomberg , kalman-filter , allocation
Unclear about this post? Read the full post at Week in Review 020212 Quantitative Finance or Asking questions and receiving
answers.
---supported by Best selling investing books
Week in Review 260112 Credit Default Swap
Time Series Matching with Dynamic Time Warping: a follow-up post for time series matching mentioned in last week.
Risk-Based Dynamic Asset Allocation with Extreme Tails and Correlations: a unique dynamic portfolio construction framework that improves portfolio performance by adjusting asset allocation in accordance with a forecast of market risk.
Problems with Using CDS to Infer Default Probabilities: banking regulations and risk management decisions should not be based on CDS implied default probabilities.
Why Borrowing Rates Should Never Be Tied to Credit Default Swap Spreads: shortfall of doing so.
Systematic Risk and the Cross-Section of Hedge Fund Returns: systematic risk is a powerful determinant of the cross-sectional differences in hedge fund returns.
Returns of the dragon: stock market returns and the Chinese zodiac.
Tags - allocation , extreme , risk , cds , hedge-fund
Unclear about this post? Read the full post at Week in Review 260112 Credit Default Swap or Asking questions and receiving
answers.
---supported by Best selling investing books
Week in Review 200112 Forecast Return
Consumer Confidence and Equity Returns: what can we learn from Michigan Consumer Confidence index to reflect future equity returns?
Time Series Matching: If history repeat itself, we can "predict" futures return. Using historical data and time series matching analysis to make an educated guess what S&P 500 will do in the next week, month, quarter. Detailed R codes are provided.
Best Practices for Programming MATLAB: List of Best Practices for Matlab coding.
Systematic Investor Toolbox: a collection of tools that we use in everyday quantitative investment research written in R.
My Life in Finance: by Eugene F. Fama.
Option Prices Leading Equity Prices: Do Option Traders Have an Information Advantage?: the answer is, not surprisingly, YES.
Trend-following and Momentum Strategies in Futures Markets: momentum trading signals generated by fitting a linear trend on the asset price path maximise the out-of-sample performance while minimizing the portfolio turnover, hence dominating the ordinary momentum trading signal in literature, the sign of past return. Second, the results show strong momentum patterns at the monthly frequency of rebalancing, relatively strong momentum patterns at the weekly frequency and relatively weak momentum patterns at the daily frequency.
Home
|
|
