Dimitris Politis
Contributing Editor Dimitris Politis writes:
Consider the standard setup where $X_1,\ldots, X_n$ are i.i.d. from distribution $F$ with mean $\mu=EX_i$, variance $ \sigma^2=E(X_i-\mu)^2>0$, skewness $\gamma = E(X_i-\mu)^3/\sigma^3$, and kurtosis $\kappa = E(X_i-\mu)^4/\sigma^4 $ assumed finite. As usual, define the sample mean $\bar X=\frac{1}{n}\sum_{i=1}^n X_i$, the sample variance $\hat \sigma^2=\frac{1}{n-1}\sum_{i=1}^n (X_i…
Dimitris Politis is a professor in the Department of Mathematics at the University of California in San Diego. He is one of the IMS Bulletin’s Contributing Editors, and a former Editor (January 2011–December 2013). Here, he writes about his most recent pastime, Model-Free Prediction:
1. Estimation
Parametric models served as…
Dimitris Politis writes: Many IMS members are educators, and thus care about improving and evaluating teaching. In the last three decades, student evaluations of classroom teaching have played a prevalent role in the US, although opinions are mixed on whether the overall effect has been positive. It goes without saying…
Dimitris Politis writes: Our journals are alive and well, and the new Annals of Applied Statistics has become an instant winner with authors and readers. As I mentioned in my letter to the IMS Bulletin (vol. 39, no. 2), the advent of manuscript-central has endowed journals with an unprecedented efficiency;…
Dimitris Politis writes: Statistics has undergone a remarkable transformation in the last two decades, adapting to a rapidly changing scientific environment. For example, under the leadership of pioneers such as the late Leo Breiman, our profession managed to prevail over challenges posed by machine learning by embracing it as the…