Computer Codes

This section is a work in progress and more functions will be added as they are prepared. However, if you would like code for any paper listed under the research section, but don't see it here, then just send me an email and I will send you what we have. All of our methods are coded in R or Matlab (sometimes with a little C++).

- AFSL and FLAME: Here is an example for FLAME (in R) and AFSL (in Matlab). These scripts can be used for variable selection in high-dimensional function-on-scalar regression, and scale quite well with the number of predictors and sample size. Note that the data for this example is simulated and discussed in more detail here.
- FMINQUE: This MATLAB function can be used to estimate the variance components in scalar, vector, and functional linear models. In otherwords, one can fit linear mixed effects models using this function. Example codes: scalar, vector, functional
- FRegion: We have an R package (authored by Hyunphil Choi) availabe for constructing confidence bands and carrying out hypothesis testing for functional parameter estimates. It is based on the work of Choi and Reimherr (2016), and is available through GitHub.
- R Binary for scpt package. This creates a function
*spatial_change_test*which carries out the spatio-temporal change-point tests described in Gromenko, Kokoszka, and Reimherr (2015). SOURCE and EXAMPLE. - A simulation example and R Code for Kokoszka and Reimherr (2013). This function tests for the order of a Functional Autoregressive Model.
- A simulation study and R Code for Horvath, Kokoszka, and Reimherr (2009). This carries out a two sample test for functional linear models.