Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

I’ve recently completed fastStat, https://github.com/matloff/fastStat,a quick introduction to statistics for those who’ve had a calculus-based probability course. Many such people later need to do statistics, and this will give them quick access. It is modeled after my R tutorial, https://github.com/matloff/fasteR, a quick introduction to R.

It is not just a quick introduction, but a REAL one, a practical one. Even those who do already know statistics will find that they learn from this tutorial.

I write at the top of the tutorial,

..many people know the mechanics of statistics very well, without truly understanding an intuitive levels what those equations are really doing, and this is our focus…For example, consider estimator bias. Students in a math stat course learn the mathematical definition of bias, after which they learn that the sample mean is unbiased and that the sample variance can be adjusted to be unbiased. But that is the last they hear of the issue…..most estimators are in fact biased, and lack ‘fixes’ like that of the sample variance. Does it matter? None of that is discussed in textbooks and courses.

The tutorial begins with sampling, with a realistic view of parametric models, estimation, standard errors, statistical inference, the Bias-Variance Tradeoff, and multivariate distributions.

It then moves to a major section on prediction, using both classical parametric and machine learning methods. Emphasis is again on the Bias-Variance Tradeoff, with a view toward overfitting. A fresh view of the latter is presented.

Finally, there is an overview of data privacy methods, of major importance today.