If you run A/B tests, you have likely encountered the term statistical power. And if you’re like me, you might know that running a power analysis before your test is the right thing to do, and that having 80% power is ‘good’. If you’re really fancy, you might have read the Wikipedia page and seen something about β, false negatives, and the like. You might have even stressed the importance of this topic to your colleagues.

Continue Reading

The other day at work I found a large bottleneck in my code. A big ETL check on millions of rows had me creating individual date sequences lots of times (don’t ask). Because its syntax was familiar and I liked how it looked, I was using R’s seq.Date to do this. Each call to seq.Date something looked like this start <- Sys.Date() end <- Sys.Date() + 10 seq.Date(start, end, 1) ## [1] "2021-07-29" "2021-07-30" "2021-07-31" "2021-08-01" "2021-08-02" ## [6] "2021-08-03" "2021-08-04" "2021-08-05" "2021-08-06" "2021-08-07" ## [11] "2021-08-08" As is the case with most decisions in programming, there are tradeoffs to using this approach, and this time the tradeoff was speed.

Continue Reading

Some statistical wisdom in these slides from Frank E Harrell at Vanderbilt. Basically reminding us that statistical validation has its limits. “We should spend more time validating investigators and less time doing ‘external’ validations”. The last plot in this blog post on dendrograms got me jazzed up. A novel visualization of “why” an observation is in a particular hierarchical cluster. Economist Branko Milanovic on why people with highly polished CV’s may not be the best social science thinkers.

Continue Reading

New hobby of mine: collecting the most memorable or noteworthy articles I catch each month. It’s a good opportunity to look back and think about news from a slightly longer timeframe. Poignant commentary may or may not appear. Chris Hughes, one of the Facebook founders, wrote a little manifesto about his former employer in the NYT. He made the case for why we should break it up, citing the grand sweeping power Zuckerberg has over the state of free speech, the press, and democracy.

Continue Reading

I was an undergraduate math major, albeit not a clever one. Proofs were hard for me, and I found it increasingly difficult to focus as the courses moved on to more and more abstract problems. An embarrassing thought: just 6 years after graduating, I don’t know if I could navigate much of my homework assignments from back in the day. But this is not to say my degree was totally in vain.

Continue Reading

Hannah Fry describes her book Hello World quite nicely in the introduction. “It’s about asking if an algorithm is having a net benefit on society. About when you should trust a machine over your own judgement, and when you should resist the temptation to leave machines in control. It’s about breaking open the algorithms and finding their limits; and about looking hard at ourselves and finding our own. About separating the harm from the good and deciding what kind of world we live in.

Continue Reading

William Gibson is too damn smart for me. Not to bore you with my reading habits, but I picked up Neuromancer back in college (over 5 years ago now, ugh) and haven’t been able to get through it after several tries. I always got lost in the density of his writing and often had to reread sentences, paragraphs, and chapters just to make sense of what was happening. Nonetheless I decided awhile back, again, to finally trudge through it all, in the hopes that completing it would help me appreciate it.

Continue Reading

The musings of a grumpy old man. Now now boy, this is where I came from. Now now boy, this is how you tell a story. Now now boy, the whole world is going to hell. And like most geezers I know, he’s got reasons to back himself up. Marriage is misery because families aren’t big enough. George Bush is “pissed off at Arabs” because they “brought us algebra”. And ole Kurt didn’t win the Nobel Prize because he trash-talked Swedish cars.

Continue Reading

Blake Mycoskie’s Start Something That Matters is quite like a pair of his famous TOMS shoes. It’s straightforward and simple, with tips telling us to “be resourceful without resources”, that “giving is good business”, and that we need to “face our fears”. The literal design of the book even echoes the blue and white tag on every pair of TOMS shoes, allowing it a consistency with the company’s mission. Mycoskie even stayed true to the TOMS One for One business model, by promising to donate one book to a child in need for every book sold.

Continue Reading

This is a very American book. Eliot Rosewater, a former socialite and the son of a senator, decides to live his life as an altruistic servant of the ordinary people in the aptly named Rosewater County. He’s an alcoholic, his wife doesn’t want him anymore, and he is mentally insane, yet Vonnegut makes us love him as a near-perfect person. At the same time, Vonnegut contrasts Eliot’s calm demeanor and prodigal generosity with a harsh depiction of just about everyone else in the book.

Continue Reading

I’m a pretty timid guy, especially when I’m around a lot of people, so I picked The Art of Public Speaking out as a self-help lessonbook of sorts. Dale Carnegie is most known as the author of How to Win Friends and Influence People–a book I have been recommended, and is on my to-read list. A quick YouTube search on Carnegie introduced me to Public Speaking as well. I had a car trip coming up so I decided to audiobook it again.

Continue Reading

I started Slaughterhouse-Five back in the summer but had a hard time getting through it, even though I enjoyed the writing and the way that it’s laid out in short passages. It probably had something to do with the fact that I don’t like books about war. So when I got to the middle of it and they were getting into the nitty-gritty of things, I didn’t feel like continuing and put the book down until recently.

Continue Reading

I listened to this one in the car on the way to Hilton Head. The length of the audiobook was about an hour less than the length of the trip, which gave me a little time by myself to think about what I had just heard. Boy, the last sentence of the book really got me. The imagery of Bokonon freezing himself, giving God the finger, was quite powerful…quite chilling in fact.

Continue Reading

LA times reported on an app that promotes political transparency—to the granularity of a bar code. I applaud BuyPartisan’s efforts to help enlighten consumers and, unlike the LA times, am not worried about the possibility of increased polarization due to apps like this. If anything, they will help citizens understand that corporations are not simply partisan, they are self-interested agents that use political donations as a means to profit maximize. A general rule of thumb: I am more concerned with the absolute dollar amount that a company devotes to politics, not necessarily its shade of red or blue.

Continue Reading

The Wikimedia blog reflected on the effect Pokemon has had on encyclopaedic knowledge in a recent post. Over its 20 year history, the Pokemon franchise has supplied us with an ever-growing heap of trivia, factoids, and backstories just waiting to be documented by the public—a cause for concern among “Exclusionists” in the Wikipedia community. I am generally on the side of the exclusionists in this matter. Plenty of well-kept wikis for fans (see Bulbapedia) exist and seem fit for this purpose.

Continue Reading

A thought-provoking study. “The researchers conclude that American universities missed out on a singular opportunity to boost mathematical knowledge because, had they accepted more doctoral students in response to the increased supply of Chinese applicants, the gains of Chinese professors would not have come at the expense of non-Chinese faculty.” To me this paper contributes to the theory that mathematical research is a highly social process; communication and the sharing of ideas are crucial to its success.

Continue Reading