Publications by John Mount
Win-Vector starts submitting content to r-bloggers.com
We have been consistently impressed by and enjoyed the wealth of R wisdom available on the R-bloggers aggregation site. Therefore Win-Vector LLC is granting the right to reformat and redistribute (with attribution and link) our blog‘s R content in the R-bloggers site and feeds. We hope to see our R content shared through this network. Related...
790 sym
Why I don’t like Dynamic Typing
A lot of people consider the static typing found in languages such as C, C++, ML, Java and Scala as needless hairshirtism. They consider the dynamic typing of languages like Lisp, Scheme, Perl, Ruby and Python as a critical advantage (ignoring other features of these languages and other efforts at generic programming such as the STL). I strongly...
9314 sym R (1028 sym/5 pcs)
Modeling Trick: the Signed Pseudo Logarithm
Much of the data that the analyst uses exhibits extraordinary range. For example: incomes, company sizes, popularity of books and any “winner takes all process”; (see: Living in A Lognormal World). Tukey recommended the logarithm as an important “stabilizing transform” (a transform that brings data into a more usable form prior to gener...
7757 sym 4 img
How to remember point shape codes in R
I suspect I am not unique in not being able to remember how to control the point shapes in R. Part of this is a documentation problem: no package ever seems to write the shapes down. All packages just use the “usual set” that derives from S-Plus and was carried through base-graphics, to grid, lattice and ggplot2. The quickest way out of th...
1269 sym 4 img
Selection in R
The design of the statistical programming language R sits in a slightly uncomfortable place between the functional programming and object oriented paradigms. The upside is you get a lot of the expressive power of both programming paradigms. A downside of this is: the not always useful variability of the language’s list and object extraction o...
14454 sym
How to outrun a crashing alien spaceship
Hollywood movies are obsessed with outrunning explosions and outrunning crashing alien spaceships. For explosions the movies give the optimal (but unusable) solution: run straight away. For crashing alien spaceships they give the same advice, but in this case it is wrong. We demonstrate the correct angle to flee. Running from a crashing alien...
4278 sym 14 img
Modeling Trick: Masked Variables
A primary problem data scientists face again and again is: how to properly adapt or treat variables so they are best possible components of a regression. Some analysts at this point delegate control to a shape choosing system like neural nets. I feel such a choice gives up far too much statistical rigor, transparency and control without real be...
9604 sym 12 img
What does a generalized linear model do?
What does a generalized linear model do? R supplies a modeling function called glm() that fits generalized linear models (abbreviated as GLMs). A natural question is what does it do and what problem is it solving for you? We work some examples and place generalized linear models in context with other techniques.For predicting a categorical out...
10145 sym 6 img
How robust is logistic regression?
Logistic Regression is a popular and effective technique for modeling categorical outcomes as a function of both continuous and categorical variables. The question is: how robust is it? Or: how robust are the common implementations? (note: we are using robust in a more standard English sense of performs well for all inputs, not in the technic...
10953 sym 6 img 1 tbl
Newton-Raphson can compute an average
In our article How robust is logistic regression? we pointed out some basic yet deep limitations of the traditional full-step Newton-Raphson or Iteratively Reweighted Least Squares methods of solving logistic regression problems (such as in R‘s standard glm() implementation). In fact in the comments we exhibit a well posed data fitting problem...
8562 sym 2 img