23 November 2020
GAM Systematic Cambridge’s Dr Linda Gruendken discusses her fascination with mathematical theory, the importance of maintaining the integrity of data and the value of ‘analogue’ thinking.
When Linda Gruendken was in graduate school in the US nearly a decade ago, there was a popular urban grad school myth that stuck with her for a long time. It was a rumour that Russian / Soviet programmers were really good compared to their American counterparts, and no one knew why. The story was that when they learned to code, they had very limited computing time available, only one to a few hours every couple of weeks.
“They would spend their working time thinking up programmes and writing them down on paper, without continuous feedback like we now have through the internet, or even with the help of syntax checks,” Linda explains, recalling the story. "They had extremely high incentives to have the whole programme written down perfectly, down to the semicolons, to make the most of their computing time. The theory was because of this, over time, they had learned to code to a much higher standard – simply because they spent so much more time thinking about their code on paper, thinking and working on an ‘analogue’ basis.”
For someone that is not an indoors person, thinking in an analogue way in her personal time, without technology, has been important for Linda during Covid-19 lockdowns. “I am certainly an outdoors person. I have always had a penchant for travelling and enjoy a variety of open-air activities, including rock climbing and bouldering. Fortunately I have travelled a lot. My trip of a lifetime would be a journey to the international space station – going to space has definitely been a lifetime dream! I am heavily focused on technology at work, so being outdoors is one way to try and keep my personal time ‘analogue’. Analogue thinking, as I call it, is time spent thinking and working without repeated, if not outright continuous, interruptions, and it is a hugely powerful creative force, even for programmers. Quality of work depends on the quality of thought.”
At work, Linda describes herself as a scientist managing systematic strategies. “I look at financial markets in a scientific way, utilising data and empirical methods. I would estimate that active portfolio tasks make up circa 20% of my working day, while 80% is spent doing research. Automating tasks is a big part of what we do at GAM Systematic Cambridge, as this allows more time for research.”
Validation of models
Linda leads the research effort into the firm’s credit strategies and was instrumental in the research and set up of the Dynamic Credit strategy, which is designed to capture upside, but also to mitigate harsh market declines, particularly in the high yield (HY) credit segment. Despite its challenges, she says 2020 has proved useful to showcase what the strategy can do. “In Q1, credit spreads reached unprecedented levels at lightning speed. The strategy was quickly able to reposition and rotate defensively, mitigating the downside. For us, it was a useful scenario and, ultimately, validation of our investment rules and processes. We place great emphasis on ensuring our investment products behave as they should. Going through a live example was an important exercise.”
Although GAM Systematic Cambridge’s systems have implied market outlooks, the company’s professionals do not spend days making up their minds regarding what may happen to markets. Linda explains: “We leave that to the models we make. That said, the outlook for credit appears bearish. Therefore it is important to have the ability to react quickly and mitigate prolonged downturns. It is also important that investors pay close attention and ensure those managing their money have sophisticated contingency measures in place. Investors should recognise that tech glitches across markets do happen and, despite planning, are difficult to prevent. I would argue systematic managers are best placed to deal with tech malfunctions, such as stock exchange outages, for example, as they can react to real time data, which is vastly different to following trend models on a spreadsheet.”
The integrity of data
We are in an age of big data, and this does not only impact investment decisions, says Linda. “Decisions across all areas of life are increasingly informed and driven by data. This means that maintaining the integrity of data against manipulation is increasingly important – partly because once your big dataset suggests a conclusion, it is actually quite hard to argue against it (‘because the data said so’). One thing that really impressed me this year is the integrity of some data scientists in the face of political bias and pressure, like Rebekah Jones in Florida. She was in charge of building a large Covid-19 database, and when she had done so, defended the datasets against being altered to support manipulated statements. She got fired for not being 'more flexible'. We all pay when data is manipulated to support incorrect solutions (in this case possibly even with our lives) – so it is important to do the right thing. I admire that kind of courage greatly.”
The poetry of mathematics
Linda obtained a PhD in 2011 from the University of Pennsylvania in ‘Higher Dimensional Class Field Theory of Varieties over Finite Fields’, which involved generalising some well-known results of number theory called class field theory to higher dimensions, using the language of algebraic geometry. She explains: “Number theory is a wonderful branch of pure mathematics, but I understand it can seem esoteric and a bit out there to some. It is a highly abstract discipline, and I often try to describe its appeal as the 'poetry of mathematics'. Just like poetry, it is a beautiful and beautifully hard discipline, and practitioners love being engaged in investigating the most abstract properties of numbers and their abstractions as a goal in itself. It has an inherent appeal to anyone fascinated by mathematical principles. But if you are immune to its charms and looking for its practical use you will not find one easily (perhaps apart from applied cryptography), again much like poetry.” Post university, Linda has focused on more empirical methods, like Bayesian statistics, and some machine learning methods. “What I am interested in is pretty strongly influenced by what might benefit our portfolio – right now I am interested in convex optimisation problems.”
Diversity is good for outcomes
Linda is always keen to teach members of her team everything she knows (in person and, more recently, virtually). “Specifically, I pass on knowledge of financial markets to our talented new joiners, many of whom are scientists. It is important for me to share knowledge of how financial markets work and what scientific models mean in relation to investors and the real world. While I am not a member of a minority and cannot call myself an expert on the topic of diversity, there is empirical evidence to suggest diversity is good for outcomes. I believe non-diverse teams tend to agree too much and can too easily devalue different opinions – this is true even if you have a scientific approach."
The information in this document is given for information purposes only and does not qualify as investment advice. Opinions and assessments contained in this document may change and reflect the point of view of GAM in the current economic environment. No liability shall be accepted for the accuracy and completeness of the information. Past performance is no indicator for the current or future development. The mentioned financial instruments are provided for illustrative purposes only and shall not be considered as a direct offering, investment recommendation or investment advice.