Posts

Kenneth Hung

**Sep 25 2020**: Playing with Bayes and RStan — I did not really much statistical training in my undergrad days, and my knowledge of statistics is pretty much confined to whatever grad level statistics classes Berkeley offered — 99% of those was frequentist — so I lack the Bayesian exposure that most statistics undergrad would have received. So when...**Aug 4 2020**: Normal distribution in Python — Working on theoretical statistics, all of my work in my PhD was done in R. But for production code in industry, for the sake of speed and easier maintenance, I have taken up to implement some of my ideas in Python even when the prototyping is done in R. I...**Feb 15 2020**: Binomial ranking with SARS data — With the coronavirus spreading to many countries, Rebecca asked me a curious question: how does the US perform during SARS compared to other regions in terms of survival rate? While we can compute the survival rate of all infected regions and rank them accordingly, we are ignoring the sampling variability....**Sep 21 2018**: p-value Screening — Will and I have been working on a problem for finding confidence lower bounds for maximum parameter. Here is a good toy model: the confidence lower bound for the maximum mean parameter for a i.i.d. Gaussian observation. Specifically, we are trying to find a confidence bound for \(\max_i \mu_i\) from...**Apr 10 2017**: ℓ1-Penalized Likelihood Asymptotics — Following a paper by Lee et al. (2013) on the correcting for the selection bias after lasso-based selection, a natural progression is to consider general penalized likelihood selections. In GLM, we are at least provided with a sufficient statistics, but this would not be the case in a more general...