James Atkins

James Atkins

Biography

I am a recent graduate of Bocconi University, Milan, where I studied for a master’s degree in Economic and Social Sciences. I gained my bachelor’s degree, also in economics, from the University of Cambridge.

I am currently working as a research assistant for Francesco Decarolis and Marco Ottaviani. Before coming to Bocconi, I worked as a consultant at DotEcon, a London based economic consultancy firm which specialised in designing and implementing high-value auctions.

Interests
  • Industrial Organisation
  • Auctions
  • Digital Markets
Education

  • MSc Economic and Social Sciences, 2018-2021

    Bocconi University, Italy

  • BA (Hons) Economics, 2011-14

    University of Cambridge, United Kingdom

Posts

Here are some things that I found useful in my work. They might be useful to you too.

Progress bar in purrr::map_df
I like purrr collection of functions rather than the base R equivalents. But when working with big datasets, it is helpful to have some indication of how long an operation is going to take.
Last updated on 10 Jan 2022 1 min read
Quantile Function of Mixture Distributions in Python
Finite mixtures are just convex combinations of two or more component distributions. Given weights $w_1, \dots, w_n$ and component distributions $P_i$ or densities $p_i$, the distribution and density functions are easy to compute, $$ F(x) = \sum_{i=1}^n w_i P_i(x) $$ $$ f(x) = \sum_{i=1}^n w_i p_i(x) $$ but the inverse distribution $F^{-1}(p)$ or quantile function is not so straightforward.
Last updated on 10 Jan 2022 2 min read