Paper2014
Tidy Data
Hadley Wickham
Defines a tidy dataset as one where every variable is a column, every observation a row and every value a cell, and shows most data-cleaning pain comes from a fixable violation of this.
link checked 17 Sept 2026Statistical Computing
Doing statistics at a scale where the arithmetic matters.
7 topics · 10 curated works
Topics
- 01Foundations & Overviews2
- 02Monte Carlo Methods2
- 03Numerical Optimisation2
- 04Simulation Studies1
- 05Bootstrap Computation1
- 06Reproducible Research1
- 07Statistical Software & Workflow1
Reading in Statistical Computing
10A way in
Start here
No prior grounding assumed.
Tidy Data
Hadley Wickham · 2014
Defines a tidy dataset as one where every variable is a column, every observation a row and every value a cell, and shows most data-cleaning pain…
+1 more at this level
Then
Assumes you know the vocabulary.
The Monte Carlo Method
Nicholas Metropolis & Stanislaw Ulam · 1949
Proposes solving otherwise-intractable integrals and diffusion problems by simulating random samples on a computer instead of solving the underlying…
+5 more at this level
Go deeper
Primary sources and full treatments.
Maximum Likelihood from Incomplete Data via the EM Algorithm
Arthur P. Dempster, Nan M. Laird & Donald B. Rubin · 1977
Formalises the EM algorithm as a general way to maximise a likelihood when data are incomplete, alternating between guessing the missing part and…
+1 more at this level
10 works
Book2017
R for Data Science
Hadley Wickham & Garrett Grolemund
Teaches the whole analysis cycle — import, tidy, transform, visualise, model — as one workflow rather than as separate tools.
link checked 17 Sept 2026Paper1949
The Monte Carlo Method
Nicholas Metropolis & Stanislaw Ulam
Proposes solving otherwise-intractable integrals and diffusion problems by simulating random samples on a computer instead of solving the underlying equations directly.
link checked 17 Sept 2026FreeIntermediate
Paper2003
An Introduction to MCMC for Machine Learning
Christophe Andrieu, Nando de Freitas, Arnaud Doucet & Michael I. Jordan
Surveys rejection sampling, importance sampling, and the main MCMC algorithms as one family of solutions to the same problem, drawing samples from a distribution too complex to sample from directly.
link checked 17 Sept 2026FreeIntermediate
Paper2011
Reproducible Research in Computational Science
Roger D. Peng
Argues that a published computational result should ship with the code and data needed to regenerate it, since peer review alone cannot catch most analysis errors.
link checked 17 Sept 2026FreeIntermediate
Paper2011
Unifying Optimization Algorithms to Aid Software System Users: optimx for R
John C. Nash & Ravi Varadhan
Compares R's competing general-purpose optimisers on a common set of test problems and shows none of them dominates, which is why optimx wraps rather than replaces them.
link checked 17 Sept 2026FreeIntermediate
Book2016
Computer Age Statistical Inference: Algorithms, Evidence, and Data Science
Bradley Efron & Trevor Hastie
Traces the arc from classical inference through the bootstrap, generalised linear models, survival analysis and machine learning as one continuous story about what computation made possible.
link checked 17 Sept 2026FreeIntermediate
Paper2019
Using Simulation Studies to Evaluate Statistical Methods
Tim P. Morris, Ian R. White & Michael J. Crowther
Sets out a checklist — aims, data-generating mechanisms, estimands, methods, performance measures — for a simulation study to be reported so another researcher can trust and reproduce it.
link checked 17 Sept 2026FreeIntermediate
Paper1977
Maximum Likelihood from Incomplete Data via the EM Algorithm
Arthur P. Dempster, Nan M. Laird & Donald B. Rubin
Formalises the EM algorithm as a general way to maximise a likelihood when data are incomplete, alternating between guessing the missing part and refitting the model.
link checked 17 Sept 2026Paper1986
Bootstrap Methods for Standard Errors, Confidence Intervals, and Other Measures of Statistical Accuracy
Bradley Efron & Robert Tibshirani
Extends the bootstrap from standard errors to bias, prediction error and confidence intervals, and works through complicated data, censored, regression, time series, where a closed-form formula would not exist at all.
link checked 17 Sept 2026In order written
1949 – 2019- 1949The Monte Carlo MethodNicholas Metropolis & Stanislaw Ulam
- 1977Maximum Likelihood from Incomplete Data via the EM AlgorithmArthur P. Dempster, Nan M. Laird & Donald B. Rubin
- 1986Bootstrap Methods for Standard Errors, Confidence Intervals, and Other Measures of Statistical AccuracyBradley Efron & Robert Tibshirani
- 2003An Introduction to MCMC for Machine LearningChristophe Andrieu, Nando de Freitas, Arnaud Doucet & Michael I. Jordan
- 2011Reproducible Research in Computational ScienceRoger D. Peng
- 2011Unifying Optimization Algorithms to Aid Software System Users: optimx for RJohn C. Nash & Ravi Varadhan
- 2014Tidy DataHadley Wickham
- 2016Computer Age Statistical Inference: Algorithms, Evidence, and Data ScienceBradley Efron & Trevor Hastie
- 2017R for Data ScienceHadley Wickham & Garrett Grolemund
- 2019Using Simulation Studies to Evaluate Statistical MethodsTim P. Morris, Ian R. White & Michael J. Crowther