This web page converts Wilkinson-Rogers notation
(used by e.g. MATLAB and R for linear models)
to a mathematical formula that can be used to calculate the interpolation in other languages, such as julia, C++ and MATLAB (for verification).
In MATLAB and R one can use for example respectively fitlm
and
lm
to create a linear regression model.
Plaese notice that the order of the coefficients given by MATLAB and R differs from the order here (sorted alphabetically).
The symbol support is listed in the following table.
symbol | example | meaning |
---|---|---|
+ | +x1 | include this variable |
- | -x1 | delete this variable |
: | x : z | include the interaction between these variables |
* | x * z | include these variables and all interactions between them |
^ | x^3 | include interactions up to three way |
1 | - 1 | the intercept (by default it is added) |
Variables