From Anna Menacher: A timeline of the most important statistical ideas of the past 50 years

Anna Menacher writes:

Over the holidays I felt a bit reflective and finally had a chance to read your paper with Aki on the most important statistical ideas of the past 50 years. When I read it I almost felt like the 8 ideas that you mentioned encompassed everything in statistics but I suppose that is only fair considering that the field of statistics only exists for about 100-200 years and you also did mention some other ideas from the last 100 years ranging from 1920 to 1970 in order to see the difference.

I found it interesting to see how the 8 influential ideas evolved simultaneously which is why I created this little timeline with the papers that you have selected.

Cool! I think it’s all the references in our paper, color-coded by topic.

5 thoughts on “From Anna Menacher: A timeline of the most important statistical ideas of the past 50 years

  1. Hmmm. Taking the references from your paper doesn’t make this “a timeline of the most important ideas of the past 50 years”, as I don’t think your paper had any ambition to put together all the really groundbreaking papers and nothing else (correct me if I’m wrong). For example I’m looking at robustness and see Stigler (2010), which is about the history of robustness and has no new statistical ideas at all in it, and a 1972 paper of Huber is cited rather than his groundbreaking 1964 one (which is correctly highlighted by Stigler but not by you).

    Maybe the headline should rather be “A timeline of all papers cited in Gelman & Vehtari (2021)”, which doesn’t raise wrong expectations?

    • Christian:

      1. It’s a timeline of the most important ideas, not the most important papers. But, yes, to be more accurate, it’s a timeline of the papers cited in a paper about the most important statistical ideas of the past 50 years.

      2. When we sent our paper to the Journal of the American Statistical Association, we suggested they publish it with discussions. For whatever reason, they decided not to. I think this was a lost opportunity. Indeed, my plan all along when writing this paper was to publish it with discussions.

    • Pass,

      It’s a 50-year summary so there’s a lot of earlier stuff. But, yeah, there’s lot of work in machine learning that Aki and I have never heard of. When we sent our paper to the Journal of the American Statistical Association, we suggested they publish it with discussions. For whatever reason, they decided not to. I think this was a lost opportunity. Indeed, my plan all along when writing this paper was to publish it with discussions.

Leave a Reply

Your email address will not be published. Required fields are marked *