5. Data are described by mathematical
models with meaningful parameters
Mathematical models are useful
for describing data
Model can grasp the form of the data from only a
few parameter values
Mathematical models are useful
for making inferences
Formal logic of mathematics allows derivation of
specific properties of parametric descriptions that
would not be obvious from the data alone
8. Thomas Bayes’ publishes his seminal
paper/theorem
1763
Probabilistic inference independently
rediscovered by LaPlace in 1774
1774
The frequentist bandwagon really gets
rolling
• Ronald Fisher developed the maximum likelihood theory
of optimal estimation
• Jerzy Neyman developed confidence intervals and tests
1900s
Computing capabilities catch up to Bayesian
thinking
New algorithms like Markov chain Monte
Carlo and Bayesian methods gain popularity
1989
Null hypothesis significance testing and p-
hacking begin to be challenged
What will you do?
2000s
10. Wasserstein, R.L., & Lazar, N.A. (2016). The ASA’s statement on p-values: Context, process, and purpose. The American Statistician, 70(2), 129–133.
http://dx.doi.org/10.1080/00031305.2016.1154108.