Chose the right tool for the right task. I‘ll go with R and RStudio or even Python for data analysis and statistics. Opting for Lisp is like trying to use a swiss knife to cut a tree just because you love your swiss knife.
First thing I did when I got my Swiss Army pocket knife was go to the woods by my house and cut down a tree with its little saw. It was a small, aspen or poplar maybe 3" thick and it took some doing but it came down. That was my first pocket knife and the first tree I cut down, believe I was in third grade. Still remember the smell of the freshly cut wood and the damp humus, the feeling of the sap running over my hand; it was one of those shadowless overcast days, early fall before leaves started turning. I avoided washing my hands all day just to keep the smell of the sap with me. I did love my Swiss knife, took it with me everywhere I went for years. Thanks for the memories.
I for instance find Python the most horrible language + ecosystem outside the js ecosystem (but I like js the language more and that's saying something), so I would always opt for lisp (or pen + paper) over Python. R / Rstudio are nice though.
I don't think it really tracks either; Lisp is quite ergonomic for this type of thing and, if you have been doing it for a while, you'll have your own tooling to work faster/more efficient in that lisp and of course, the comparison falls down then as the swiss knife now has a chainsaw option which is as good or better than other options to cut down trees.
Reminder that before Python was used for data science, people used things like BioPerl and PDL and that didn't stop people from working on pandas and the like.
Also let people have fun.
So Julia is a happy middle ground - MATLAB-like syntax with metaprogramming facilities (i.e., macros, access to ASTs). Its canonical implementation is JIT, but the community is working on allowing creation of medium-sized binaries (there has been much effort to reduce this footprint).
1. readability with explicit broadcast operators
2. interoperability with other languages including R and Python
3. performance often exceeding numpy and C/C++ code
4. usability in numerous workflows:
The idea of using Lisp or Prolog in a production environment doesn't sound fun at all. Yet, they do make some types of problems easier to handle. =3