This post is not about politics, but the failure of statistics applied to politics. Our technology, the science behind big data, Bayesian statistics, and ultimately their incarnation in Nate Silver's model, all crumbled down last night. Sure, the FiveThrityEight blog keeps saying that there was a high level of uncertainty associated with the polls, mostly due … Continue reading The Scariest Thing of All
In a previous post I used a dataset taken from my Facebook friends to debunk the claim that more babies are born under a new moon. I have now a chance to increase the test statistics significantly, not because in the meanwhile I made many more friends, but because people collaborating to the FiveThirtyEight blog … Continue reading Are more babies born under a new moon? – pt.2
The big news in particle physics during the winter 2015-2016 is certainly the excess found by the ATLAS and CMS Collaborations in the invariant mass of two high-energetic photons in the data acquired in 2015 at a centre-of-mass energy of 13 TeV. If you don't know what I'm talking about, then it's time to catch … Continue reading A few thoughts about the 750 GeV diphoton bump
I'm reading a very interesting book written by Seth Lloyd called "Programming the Universe: a quantum computer scientist takes on the Cosmos". Highly recommended! I enjoyed in particular how entropy is compared to a spreading disease: a disease of ignorance. What is it all about? Let me explain briefly. From statistical mechanics, entropy is a measure … Continue reading A Little Bit of Entropy
Introduzione Visto che le prossime elezioni si avvicinano, è bene prepararsi per tempo… Con questo post voglio fornire un semplice strumento di calcolo per effettuare in casa le proiezioni elettorali. Sia chiaro, non ho la pretesa che sia un metodo accurato: in questa prima versione la cosiddetta forchetta o forbice è data semplicemente come incertezza … Continue reading Te le do io le proiezioni!
In brief: a moderate YES! This time, I analyzed 1000 tweets every 20 seconds. Below you can see the raw data (grey) and the 1-minute time-averaged values. Once the fluctuations are smoothed out, the trend is clearly positive.