Merge pull request #203 from tidymodels/develop

v0.4.0

Author: andrewpbray

.Rbuildignore

@@ -1,3 +1,4 @@
+^CRAN-RELEASE$
 ^.*\.Rproj$
 ^\.Rproj\.user$
 ^README\.Rmd$
@@ -18,3 +19,5 @@
 ^TO-DO\.md$
 ^\.httr-oauth$
 ^_pkgdown.yml
+^_pkgdown\.yml$
+^docs$

.travis.yml

@@ -19,7 +19,7 @@ notifications:
 
 r_packages:
   - devtools
-  
+
 r_github_packages:
   - hadley/pkgdown
   - tidymodels/infer
@@ -35,9 +35,6 @@ deploy:
   github_token: $GITHUBTRAVIS
 #  target-branch: gh-pages-dev
   target-branch: gh-pages
-  on:
-#   branch: develop
-    branch: master
 
 after_success:
   - Rscript -e 'covr::codecov()'

DESCRIPTION

@@ -1,20 +1,21 @@
 Package: infer
 Type: Package
 Title: Tidy Statistical Inference
-Version: 0.3.1
+Version: 0.4.0
 Authors@R: c(
     person("Andrew", "Bray", email = "[email protected]", role = c("aut", "cre")),
     person("Chester", "Ismay", email = "[email protected]", role = "aut"),
     person("Ben", "Baumer", email = "[email protected]", role = "aut"),
     person("Mine", "Cetinkaya-Rundel", email = "[email protected]", role = "aut"),
+    person("Evgeni", "Chasnovski", email = "[email protected]", role = "ctb"),
     person("Ted", "Laderas", email = "[email protected]", role = "ctb"),
     person("Nick", "Solomon", email = "[email protected]", role = "ctb"),
     person("Johanna", "Hardin", email = "[email protected]", role = "ctb"),
-    person("Albert", "Kim", email = "[email protected]", role = "ctb"),
+    person("Albert Y.", "Kim", email = "[email protected]", role = "ctb"),
     person("Neal", "Fultz", email = "[email protected]", role = "ctb"),
     person("Doug", "Friedman", email = "[email protected]", role = "ctb"),
     person("Richie", "Cotton", email = "[email protected]", role = "ctb"),
-    person("Evgeni", "Chasnovski", email = "evgeni.chasnovski@gmail.com", role = "ctb"))
+    person("Brian", "Fannin", email = "captain@pirategrunt.com", role = "ctb"))
 Description: The objective of this package is to perform inference using an expressive statistical grammar that coheres with the tidy design framework. 
 License: CC0
 Encoding: UTF-8
@@ -26,7 +27,8 @@ Imports:
     rlang (>= 0.2.0),
     ggplot2,
     magrittr,
-    glue
+    glue (>= 1.3.0),
+    grDevices
 Depends:
     R (>= 3.1.2)
 Suggests: 

NAMESPACE

@@ -2,20 +2,27 @@
 
 S3method(print,infer)
 export("%>%")
+export(GENERATION_TYPES)
 export(calculate)
 export(chisq_stat)
 export(chisq_test)
 export(conf_int)
 export(generate)
 export(get_ci)
 export(get_confidence_interval)
+export(get_p_value)
 export(get_pvalue)
 export(hypothesize)
 export(p_value)
 export(rep_sample_n)
+export(shade_ci)
+export(shade_confidence_interval)
+export(shade_p_value)
+export(shade_pvalue)
 export(specify)
 export(t_stat)
 export(t_test)
+export(visualise)
 export(visualize)
 importFrom(dplyr,bind_rows)
 importFrom(dplyr,group_by)
@@ -36,6 +43,7 @@ importFrom(ggplot2,ggtitle)
 importFrom(ggplot2,stat_function)
 importFrom(ggplot2,xlab)
 importFrom(ggplot2,ylab)
+importFrom(glue,glue_collapse)
 importFrom(magrittr,"%>%")
 importFrom(methods,hasArg)
 importFrom(rlang,"!!")

NEWS.md

@@ -1,3 +1,32 @@
+# infer 0.4.0
+
+## Breaking changes
+
+- Changed method of computing two-sided p-value to a more conventional one. It also makes `get_pvalue()` and `visualize()` more aligned (#205).
+
+## Deprecation changes
+
+- Deprecated `p_value()` (use `get_p_value()` instead) (#180).
+- Deprecated `conf_int()` (use `get_confidence_interval()` instead) (#180).
+- Deprecated (via warnings) plotting p-value and confidence interval in `visualize()` (use new functions `shade_p_value()` and `shade_confidence_interval()` instead) (#178).
+
+## New functions
+
+- `shade_p_value()` - {ggplot2}-like layer function to add information about p-value region to `visualize()` output. Has alias `shade_pvalue()`.
+- `shade_confidence_interval()` - {ggplot2}-like layer function to add information about confidence interval region to `visualize()` output. Has alias `shade_ci()`.
+
+## Other
+
+- Account for `NULL` value in left hand side of formula in `specify()` (#156) and `type` in `generate()` (#157).
+- Update documentation code to follow tidyverse style guide (#159).
+- Remove help page for internal `set_params()` (#165).
+- Fully use {tibble} (#166).
+- Fix `calculate()` to not depend on order of `p` for `type = "simulate"` (#122).
+- Reduce code duplication (#173).
+- Make transparancy in `visualize()` to not depend on method and data volume.
+- Make `visualize()` work for "One sample t" theoretical type with `method = "both"`.
+- Add `stat = "sum"` and `stat = "count"` options to `calculate()` (#50).
+
 # infer 0.3.1
 
 - Stop using package {assertive} in favor of custom type checks (#149)