Initialize Travis;

Add cran-comments.md file to record CRAN notes;

Fix errors and warnings when building PAsso locally and on Travis;

Fix diagnostic.plot documentation;

Clean up dependents to speed up loading of the PAsso;

- Update README and other files.

Update “PAsso()” to support adjacent categories regression model by using vglm(acat());

Update “residuals()” to support acat in VGAM;

Remove useless dataset df_AdjCat; functions “gof”;

Rename attribute “boot_reps” as “draws”.

- The wolfsigma method of correlation works now even when n_draws>1.
- A issue when n_draws=1 in residuals.PAsso(), save array even if n_draws=1.

Add ‘diagnostic.plot’ to replace ‘check.model’ for the diagnostics of fitted models in PAsso object. ‘autoplot’ become internal S3 method.

S3 ‘diagnostic.plot’ method can deal with classes of PAsso, PAsso.test, resid, clm, glm, lrm, orm, polr.

Only keep one nes2016 dataset.

Change ‘resids’ function to a S3 method ‘residuals’, which is consistent with R core ‘stats’ package.

Conduct more test by ‘testthat’ to avoid bugs when methods are applied to different classes.

- Jittering on the probability scale is currently only supported for logit-type models.
- issues of ‘plot’ and ‘diagnostic.plot’.

Replace ‘cor’ function by ‘pcaPP::cor.fk(X)’ to speed up the PAsso().

Finalize the S3 methods ‘summary.PAsso’ and ‘print.PAsso.test’ to provide clean results.

- The digits in the print and summary are consistent.
- The manual is updated and S3 ‘autoplot’ is shortened.
- check.model for qq-plot is fixed.

- Change the name of package to “PAsso”.
- Change the autoplot function as a S3 method.

Rename functions: corr -> PAsso; corr.test -> test; ggpairs -> plot; checkmodel -> check.model.

Update print and summary functions for PAsso and test.

Add S3 function “plot3D” to draw 3D P-P plot.

Add nes2016_raw, nes2016_pre, nes2016_post, nes2016_prepost datasets.

Include nes2016.clean() function to clean up nes2016_raw dataset.

- Add
`corr`

and`corr.test`

function to estimate and test the partial correlation coefficients; - Update S3 generic functions (print, summary) for the class “PAsso” generated by “corr”, and “PAsso.test” generated by “corr.test”.

- “polr” and “glm” functions work.
- The formula in the model can be shown by the print(“PAsso”).