Hi, I’m Chris. Welcome to my blog.

About Quantitative Thinking

We all know the famous saying: “Through experience we learn”. In my view, rigorous quantitative thinking is just a way to exploit this to the fullest. Two important aspects associated with a quantitative approach seem particularly well-suited to create some added value. First, there is no reason why we should learn from our own experiences only. Instead of smoking ourselves for 20 years in order to find out about the adverse long-term effects on health we just might look at the experiences of other people that already have carried out this unfortunate experiment on their own. In other words: we can simply rely on a broader database, filled with experiences from multiple persons, instead of only relying on things that we have experienced ourselves. And second, quantitative thinking also might improve the efficiency of learning from any given experience. Overestimation, overconfidence, biases - there are many human traits that can get in the way of objective evaluations and hence inhibit optimal decision making. Using a rigorous quantitative modeling approach reduces these deficiencies and puts decision making into a testable and verifiable framework - a prerequisite to evaluate success. With the right tools, data can reveal insights that go far beyond the history and experiences of any single person.


More concretely, the following components are essential to a quantitative approach:


  • Quantification: Instead of talking about “nice” or “hot” weather, it is much more precise to compare days in terms of objective measures like temperature or humidity.
  • Data: data enables us to learn from past events. But of course, “with great power comes great responsibility” - we need to be very cautious about how we use, store and interpret data.
  • Models: Combining data with well thought out assumptions allows to draw inferences beyond what we already can see in our data. For example, assuming independence between coin tosses and a probability of $0.5$ for tails we can simply infer the probability of 100 consecutive times tails as $0.5^{100}$.
  • Visualizations: A picture is worth a thousand words. In many situations well-suited visualizations will be the easiest and fastest way to discover patterns in data, summarize findings and convey a certain finding to other people.
  • Tools: Whether it is storing and accessing of data, inferences from complicated models or visualizations - none of these steps is feasible in practice without an appropriate software setup. Hence, software tools will also be part of the topics in this blog.

About Me

My enthusiasm for statistical modeling goes back many years. I graduated in financial mathematics before I became a research assistant at the statistics department of Ludwig-Maximilians University (LMU). During that time I wrote my doctoral thesis at the chair of financial econometrics on “Dynamic Risk Management of Multi-asset Portfolios”.


I’m generally interested in a broad spectrum of socioeconomic topics, for example:

  • Earth observation, GIS
  • Smart city planning
  • Economics, Finance
  • Crypto


My goal has always been to achieve a holistic view on socioeconomic phenomena, and to build a strong understanding about statistical methodologies. I am convinced that the best way of getting there is through active engagement with real world data, using statistical / AI modeling and state of the art software tools. In addition, I also like data visualizations and 3D modeling with Blender 3d. You know, to make sure to live up to the claim that “information is beautiful” ;-)


If you want to get more details about me, here is my CV.