Author Archives: quantuniversity

QuantUniversity did a presentation on Model Risk Management at the Data Financial for Financial services conference in Boston

We presented a talk on a “Framework for Model Risk Management” at the Data Financial for Financial services conference in Boston on July 22nd 2014.

The slides of the presentation can be found here:

 

QuantUniversity’s article on Best Practices for Model Risk Management Published in the Wilmott magazine

decalogueQuantUniversity’s article on Best Practices for Model Risk Management Published in the July 2014 edition of the Wilmott magazine.

decaloguedecalogue

An early copy can be downloaded here: http://www.quantuniversity.com/w9.html

The official copy can be downloaded at: http://onlinelibrary.wiley.com/doi/10.1002/wilm.10348/abstract

Summary:

Model risk and the importance of model risk management has gotten significant attention in the last few years. As financial companies increase their reliance on quants and quantitative models for decision making, they are increasingly exposed to model risk and are looking for ways to mitigate it. The financial crisis of 2008 and various high profile financial accidents due to model failures has brought model risk management to the forefront as an important topic to be addressed. Many regulatory efforts (Solvency II, Basel III, Dodd-Frank etc.) have been initiated obligating banks and financial institutions to incorporate formal model risk management programs to address model risk. Regulatory agencies have issued guidance letters and supervisory insights to assist companies in developing model risk management programs. In the United States, as the Dodd-Frank act is implemented, newer guidance letters have been issued that emphasize model risk management. Despite these efforts, in practice, financial companies continue to struggle in formulating and developing a model risk management program. A lot of companies acknowledge and understand the model risk management guidelines in spirit but have practical challenges in implementing these guidance letters. In our prior article on model risk , we discussed many drivers to address model risk and challenges in integrating model risk into the quant development process.

In this article, we will discuss ten best practices for the implementation of an effective model risk management program. These best practices have evolved from discussions with industry experts and consulting projects we have worked with in the recent years to create robust risk management programs. These best practices meant to provide practical tips for companies embarking on a formal model risk management program or enhancing their model risk methodologies to address the new realities.

Link

Fig1

Interest in Big data technologies has significantly grown in the last few years. Data growth has exploded in the last decade. Information systems are generating huge amounts of data, trading data is becoming much more granular and at higher frequency, unstructured data from Twitter feeds, blogs etc is becoming mainstream and technologies to produce, transmit, store and process these massive datasets is collectively enabling the Big data revolution. Data science is the one of the hottest areas for internet startups. Cloud computing is making access to unlimited hardware at your fingertips. Significant venture capital money is flowing to support newer and innovative technologies that leverage vast amounts of data to create platforms and applications that could enable decision making close to the proverbial “speed of light”. Technology industry stalwarts like Google, Microsoft, Amazon and Oracle are making significant investments in Big data technologies. The applications and opportunities seem endless and the promise of revolutionary applications in healthcare, finance, space research and pure science has led various industry and academic organizations to significantly ramp up efforts to develop technologies to help analyze massive data sets. Quants have been in the forefront in the financial industry in adopting revolutionary technologies and the Big-Data phenomenon has undoubtedly caught interest in the quant community. In every quant gathering, there is at least one reference to Big data and discussions on the potential and possibilities. Quants have started to frequent Big data conferences and events to know more about technologies and opportunities in this space. I have had multiple discussions with clients who want to start Big data projects but many still struggle to understand what it is about and how they can leverage these technologies to further their quantitative research ideas. The information overload on Big data and the umpteen numbers of sources of Big data, primarily vendor and media driven, is adding to the confusion leaving quants to ponder on whether Big data is another fad or is there a real opportunity they should try out to gain an edge. I started out to write an article to introduce Big data but rather than just rehash information that is widely available, I thought it may be useful to share the my perspective as a quant practitioner with experiences from the analytics and the quant worlds on pointers to help quants understand the realities before embarking on a Big data project. In this article, I will start out with a brief introduction to Big data and point you to sources where you can learn much more about Big data. I then discuss five key things quants should consider before embarking on your first Big data project. I will then conclude with pointers on how quants can keep themselves in tune with the rapid innovations happening in the Big data world.

Download the article at http://www.quantuniversity.com/w7.html

Two new courses to be taught in Spring 2014

We will be teaching 2 new semester long courses in Spring 2014.

  • Business Intelligence, Analytics and Visualization  at Babson College: A semester long course for MBA graduate students 

BabsonLogo

Summary:

This course will examine the methods and challenges faced in turning data into insightful analytics in business. With data sizes significantly increasing in the last decade, extracting meaningful information to compete successfully is essential. You will accomplish this by learning techniques for data gathering, data analysis, and visualization as well as in discussion on companies currently trying to turn the information they gather into business opportunities. We will learn a variety of methods and software for finding patterns(such as regression, neural networks, association rules, CART, forecasting etc.), building models, and ultimately making decisions using large data sets. This is a hands-on course with in-class exercises and group projects to help students learn and apply data analysis techniques preparing them for the practical challenges analysts face in the real world.

  • Data Science and Analytics  at Northeastern University : A semester long course for graduate students in the  MS in Information Systems program at Northeastern University

neu-seal

Summary:

This case-study based course introduces various data science and analytical techniques. We will introduce various machine learning techniques and work with data sets to understand how these techniques can be applied to turn data into knowledge. Case studies from retail, energy, finance and marketing domains would be used to motivate the understanding of data science techniques.   This is a hands-on course with in-class exercises and group projects to help students learn and apply data analysis techniques preparing them for the practical challenges analysts face in the real world.