Meet our Financial Math Alumni- Coffee chat with Albert Hopping- Part 2


Part II: Analytic techniques continued...

6) Do you believe the future movement of the market data is, to some extent, driven by the models used by major financial institutions, even though these models may not be correct?

Albert: Does the market move because the model is assumed to be correct? I would say yes. Consider the market crash in 1987. Prior to the crash, there was no “volatility smile,” but there was after the crash. The market had actually been acting correctly, according to the model, until they realized the model made terrible assumptions. The market moved much more than their model implied. One event is meaningless; however, this highlights an incorrect model driving the market.

For a more current example, look to the way mortgages were priced before the recent mortgage bubble. Banks priced mortgages with the assumption that housing prices wouldn’t fall. The entire market priced that way because that was what everybody else did; it was group think. They used this assumption because the price had never gone down in their historical data set. Models are only as good as their assumptions. Could somebody have built a better model and actually predicted the housing price collapse? Yes, it could have been done and a few people did it.

Unfortunately, in a time of high earnings, it is easy to ignore risk. Risk is especially underweighted when quarterly earnings are prioritized over long term security. I feel that compensation packages are lagging the culture shift at most companies. This disconnect leads people to act as individuals focused on their personal bonuses rather than acting as representatives of their company. Further, a system which provides bailouts for bad behavior begets that bad behavior. The “smart” companies wouldn’t receive a bailout, leading some to assume they are better off to employ the incorrect group think model. Of course, this would not work in a free market.

7) Does your company use stochastic models? If it does, what kind of models are used?

Albert: We primarily work with customers who have reasons for the models they use. Sometimes, they ask us to implement specific models in their system. Other times, they come to us for advice asking what model may be best. For instance, in regards to interest rates, I am a big fan of the Libor Market Model (also known as the BGM Model). It is a term structure model. However, the industry almost exclusively seems to use short rate models. The Hull-White Model is very common because it’s very easy to parameterize, simple, and everyone else is using it.

However, I enjoy commodities more than interest rates or equities. In commodities, we have different issues because it is such a physical market. These models can be much more complicated. If I picked a favorite model, it is one that I was lucky enough to have helped develop (I’m biased). That model takes the term structure for a commodity and relates it back to the spot price allowing them to be simulated together. I really enjoy working with that model. In general, my favorite model is the right model for the situation; a model that makes logical sense and fits the data.

Part III: Risk management

8) How do the recent financial crisis and the regulation policies enacted after that affect the behavior of your company?

Albert: As a vendor, our business is based on our customers’ business. A crisis like that causes additional regulation or at least the changing of regulation. To handling that regulation it is very logical that a third party, a vendor would make a solution and sell it to customers. Typically, that type of regulation would cause a company like SAS to make a new product and be able to provide that solution to more people. Unfortunately, the capital expended on satisfying regulations cannot be used elsewhere in the economy.

9) The goal of risk management is to achieve a balance between returns and risks. Thus, with lots of capital and human resource spent, risk management may, to some extent, reduce a company’s profits. Now suppose you are a leader of a financial institution. Driven by the motivation of maximizing the profits, will you pay enough attention for risk management?

Albert: As a risk professional my answer must be yes. There are two aspects to consider in regards to risk: monitoring and management. Consider risk monitoring first. One should spend resource and pay attention to know the rules of the game. For an example, let’s think back to mortgages. What if housing prices could go down? I may go bankrupt. Well, that would be very important to know. If you don’t know the rules of the game, you cannot play your best.

In the same way, you need risk monitoring to help you see what the possibilities are. In terms of risk management, it’s like getting an insurance policy. Let’s say I have a house and all my money is in my house. If my house burns down, I may go bankrupt. I should clearly buy fire insurance on my house. That’s how risk management can be considered as well. Yes, if I am concerned about anything other than the very short term, I would spend enough resource on risk management and pay attention to my risk team.

Part IV: Suggestions & Advice

10) What skill sets are important to succeed in your field?

Albert: I get asked that question often: by students, by some of my friends, and some of my peers. I change the details of my answer almost every time, but there are key components that remain consistent. First and foremost is communication. No matter how smart you are, no matter how brilliant your model may be, if you cannot convince others and if you cannot explain your ideas, it is not going to matter.

Another component in my list is passion. Passion is not a skill, but an ingredient to success. Is it necessary? No. But if you are not passionate about your field, why work in it? Your passion allows you to have better ideas and think outside of the box which is critical. If you just think like everyone else, you are very replaceable. This makes it more difficult to advance. Your passion may present itself in the form of problem solving. This is an important skill.

Another skill of importance is programming. In our field, people are often not formally trained in programming. We are more likely to be self-taught with at most one or two university courses in programming. This is very different than those people who come out of school with an entire degree in computer science. They have different level of understanding the way a machine thinks. In some parts of our field, this understanding is critical and you will need to learn it. Being skillful enough to have a computer automate you work is always important. Automation frees your time to think and add real value.


Thank you Albert for inviting us to SAS for coffee and taking the time for this fun and informative interview!


Meet our Financial Math Alumni- Coffee Chat with Albert Hopping- Part 1


Meet Albert Hopping, ERP- Manager of Risk Consulting at SAS Institute in Cary, North Carolina. Albert graduated from the Financial Math program in 2007. He is a Board Member and an active alumnus of our program. We were happy to meet with Albert for coffee at his office at SAS.

Part I: Education & Job Background:

1) Why did you decide to get a Master’s degree in Financial Mathematics from North Carolina State University?

Albert: That I might receive more wages. That is the short answer. I finished my undergraduate degree before this program existed. When I joined the program, I was working full time. I had been out of my undergraduate program for a few years and I ended up in a risk analytics team at a diversified energy company. At that point, I didn’t have a risk background and didn’t know much about the field. I was on the risk team looking around and I saw these quant guys programming in Matlab. It looked like fun to me and I thought that it was a cool job. I wanted to be a part of what they were doing, so I started helping them with their work as much as I could.

Eventually, it got to the point where I was doing this risk work a majority of my time. I told my manager that I should be moved to the quant job family. I was told that a master’s degree in Financial Mathematics was a prerequisite for the job. Not having the degree was a roadblock for me. I applied to the Financial Mathematics program that week and I am glad I did.

2) How did the program prepare you for your job?

Albert: I was already doing risk work; I was self-taught to a certain extent. This put me in a different position compared to most students, and I got different things out of the program than other people might have, as a result. Most students learn theory first and practice second. I started the program with the perspective of a practitioner. While in the program, I learned about models I used on the job. I used these tools at work, but I didn’t really know about stochastic partial differential equations. I used Black-Scholes, but I could not derive it.

What I received from the classes is a much deeper conceptual understanding and a firmer foundation from which to see my work. What I really took from this program is a fundamental, basic understanding of financial mathematics. I also learned new models and conceived great ideas to use in practice.

3) Please briefly describe your job, your job title, and your responsibility?

Albert: At SAS Institute, I am a Manager on the Risk Solution team of Professional Services & Delivery (PSD). Let me explain from the top down. PSD is the consulting, customization, and delivery arm of SAS. Many customers of SAS software want services, consulting, or even staff augmentation. PSD provides these services. We are the ones who go to the customer site and work with the customer to help them get the most benefit from our products. Our team within PSD specializes in the risk management domain. We work with all the risk solutions and provide consulting for all risk topics. Our team has about 20 members and is growing.

Within risk, I specialize in three industries Energy, Financial Services, and Healthcare. I am responsible for leading customer projects, providing industry and risk domain expertise to the sales teams, mentoring fellow team members, and most importantly providing value to the customer. Note that the views and opinions I express today are my own.

Part II: Analytic techniques

4) “Big Data” is a hot specialization in the field. Do you see this as a long term trend or something that might pass as a fad?

Albert: Big Data is definitely a long term trend. In fact, I would go beyond that; I would say it is going to be the new norm. It will progress to the point where big data is simply the paradigm. I will even extend that to unstructured data. Companies, who are not using big data and unstructured data to their advantage, are starting to fall behind. They are tomorrow’s luddites.

5) The trend of “Big Data” implies that people believe historical data can shed light on future prediction. However, this contradicts with “efficient market hypothesis” to some degree. What are your thoughts about this?

Albert: One of the things I would like to point out in terms of the “efficient market hypothesis,” is the irrationality in the market. A simple example comes to mind: technical traders discuss how a stock index will meet resistance or break through a barrier. But what are those points where the index meets resistance or breaks through? They are numbers with lots of zeroes on the end, round numbers. Why are those numbers important? They are only important because we tend to be emotional and we have ten fingers. I propose that if we had a different number of fingers we would use a different base for our number system. The round numbers where the stock index meets resistance would be different numbers.

Clearly, these barriers are irrational, as they are based on how many fingers we have. This means I cannot be a full believer in the “efficient market hypothesis.” The question remains, is all this historical data priced into the market already? To the extent that people are doing analytics on big data, perhaps yes. Was it priced in before? No. Was the data available? Mostly, but people could not convert the data into knowledge. It was impossible - until analytics on this big data was possible.

Now, we are in the place where something can be done because of the advancements in software and the physical hardware. Data can be restructured and put into use in the market. The fact that the data is available is clearly important, but prior to these advancements one could not glean actual insights. The data must be converted into information that helps those insights that yield a better price or a better model. Acting upon those insights is what makes the market more “efficient”.


Stay tuned for Part 2 & follow Albert on twitter @SASQuant