Economics of whatever

In my line of work as a strategy consultant, I sometimes work on techno-economic studies. Using a combination of statistical analysis, forecasting techniques and calculations, we estimate various different cost trajectories, and examine the economics of something. It could be a project, a technology, or a decision that a company is trying to undertake which has some cost impact and some benefits somewhere else.

In economics, we can only perform estimates when we assume all else equal. That’s the only way to perform proper sensitivity analysis. When you change more than one parameter, then you’d call it a scenario analysis. There are infinite possible scenarios when there is infinite parameters to shift or parameters with continuous range of possibilities. So when we model the economics of something, we’d always be varying something that we think we have control over, or that we expect will be changing in the near term, and see the effects on the economics.

Yet we often think that the economics of a technology or something will remain the same unless something drastic happens. More often than not, economics of technologies shift when various players in the market make different investments, either in the underlying technologies, manufacturing capacity for the gears and components, or simply into research and development. The actions of many different parties can have a collective impact of making the economics work when at first it doesn’t seem to work.

When we critique the economics of a new innovation or project, we often forget we are comparing them against the status quo where many are already very well-invested into. The years, decades and even centuries of using a technology, manufacturing it, creating complex supply chain and auxiliaries around the status quo. It is naturally hard to beat. But what is critical about a new technology is that the incremental investments can make a large impact, small changes to scale can also make a difference. Coordination and changing expectations play a big role.

Will the economics of new innovations change overnight? Unlikely, but they typically change faster than you and I can work out the math for the economics of it.

Coursera Data Science

So for the past few months since August, I’ve been sprucing up on my skills in R through the Coursera online course designed by a couple of folks from the John Hopkins Bloomberg School of Biostatistics. It’s been interesting to pick up not only data analytics skills but learning about the different tools and platforms available online just for presenting data, doing visualizations and all that fancy stuff. I find it amazing how I first learn about Bootstrap, Spline Regressions, Permutation Tests, and a couple of other statistical techniques that might be useful in Econometrics through some of those courses in the Data Science Specialization.

So shortly after picking up those cool skills and programming some rather sophisticated looking data processing stuff (at least in my opinion), I actually started presenting stuff on various platforms using R! I wrote my first data analysis report on Rpubs, started a Github account and picked up Git (just a little; I simply wasn’t geeky enough), developed this rather useless fancy webapp and then even pumped out somewhat cooler slides to pitch it. All in all, I believe I gained more in this 3 months of learning from the series of online course than stuff I picked up in school over the past year (granted, that Advanced Math course sure was tough and made me feel like I picked up something, though I promptly forgot all of them).