This is a package for calculating confidence bands on a rectangular region. KB always stands for confidence bands and the rectangular region is usually from the minimum to the maximum of the regression model.
This is the uploaded version of my bachelor thesis. This R package calculates confidence bands with the application of stem cell data, which is included in /data, in mind. There is also a test routine to calculate the coverage probability of the confidence bands.
This is done in a multistep process. Firstly, the data is read and saved in /data as an .rda file. Furthermore, the model, notably the design matrix, is created and fixed values, like alpha, are initialised. All of this is done in the files /R/make-test-data-R.R, /R/make-test-data-R-pruef.R, /R/make-test-data-AR.R, /R/make-test-data-AR-pruef.R for the coverage probability and in /R/convert-stem-data.R for the stem data. However, for the stem data there is no central point where the fixed values are initialised.
Then the following steps are performed:
Then there is a fifth step to create a plot or to calculate a coverage probability.
Details for each numbered step above are now discussed. Furthermore, notice that each step has a corresponding R script in /R.
The fifth step of giving out the result is usually done in a two step process. First, there is a function in /R which is one of
The first two calculate the coverage probabilty for the different methods and models. The others create plots for the stem cell data.
For the coverage probability functions there is an R script in /tests that calls the functions above with the right parameters. The same is true for the stem cell data with the difference, that the calling script is in /man.
There are a bunch of supplementary functions. Two important ones are support functions in R/support-functinons.R and a test evaluation function in /R/Test-function.R.
The main problem with the coverage probabilities is, that the number of iterations in the density simulations is way too low. This is due to running time problems. These problems in turn lead to the creation of the function KB.minmax.poly.fast, but the issue still persists.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.