Katie's Tasks: - Work on residual vignette (IN PROGRESS) - include a link to the residual document in the documentation - modify so works for predictive models (Research at least)(NOT FOR CRAN) -randomForest (from randomForest): contains 'y' and 'predicted' (class is randomForest.formula)( - neuralnet (from neuralnet): contains 'response', 'covariates' and 'data' (class is nn) - nnet (from nnet): contains 'residuals' and 'fitted.values' (class is nnet) - svm (from e1071): contains 'fitted' and 'residuals' (class is svm) - test file using testthat - how to add lmertest to testthat, but still test lmer as well ??? - Added in section for warnings and errors, tested all formatting options with 2 plots - NOTE:USING THE WORD plot instead of plots - NOTE: if plotting a panel of four plots, nrow DOES NOT USE THIRD ROW when nrow = 3 - round all values shown in interactive plots - Check resid and predicted
shorten title of constant leverage plot
fix resid_interact for binomial case to be more general (Double check names with Katherine)
Katherine's Tasks - current:
Ideas for Future Package Versions: - write a function for computing standardized residuals for lmer and glmer (Katie) - look into testthat and vistest packages for testing - create an option for the user to add their own theme - add in minimum package versions?? (under imports) - add interactive versions of the auxpanel, xpanel, and compare plots - maybe add an option in each of the functions for an interactive version - option to color residuals by predictor variables? - add a resid versus y plot - read about mixed model influence diagnostics in SAS for mixed models (Stroup 2018) - nested residuals/plots for assessing random effects
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.