Computing for Statisticians
Statistical computing is the practical use of software to do statistics, mainly R and Python with pandas, NumPy and SciPy. It covers importing and cleaning data, simulation, visualisation and reproducible reports. Simulation and resampling methods such as the bootstrap let you check theory by running experiments on a computer.
Stage: Foundations · Part of the Statistician roadmap · Last reviewed 19 November 2026
What to learn in Computing
- Load, clean and reshape a dataset in R with the tidyverse or in pandas
- Draw random samples and simulate a sampling distribution
- Plot distributions and relationships with ggplot2 or matplotlib
- Implement a bootstrap confidence interval for a median
- Produce a reproducible report with a script or notebook
Computing interview questions
What is the bootstrap, and why is it used?
The bootstrap repeatedly resamples your data with replacement and recomputes a statistic each time, giving an empirical picture of its variability. It lets you build confidence intervals for statistics, such as a median, where a simple formula is hard or assumptions are doubtful.
Why set a random seed in a simulation?
A seed fixes the pseudo-random number generator's starting state so the same code produces the same draws. This makes results reproducible for review and debugging. Different seeds still show how much results vary by chance, so check more than one.
Where Computing fits in the Statistician roadmap
Computing belongs to the Foundations stage (beginner level), which takes about 50 hours. Probability and data.
Other units in this stage
Back to the full Statistician roadmap