Merge pull request #36 from jhelvy/rc-v0.7.0

Rc v0.7.0

Author: jhelvy

CRAN-RELEASE

@@ -0,0 +1,2 @@
+This package was submitted to CRAN on 2022-06-16.
+Once it is accepted, delete this file and tag the release (commit f56b55c).

DESCRIPTION

@@ -1,6 +1,6 @@
 Package: logitr
 Title: Logit Models w/Preference & WTP Space Utility Parameterizations
-Version: 0.6.1
+Version: 0.7.0
 Authors@R: c(
     person(given   = "John",
            family  = "Helveston",

NEWS.md

@@ -1,3 +1,13 @@
+# logitr 0.7.0
+
+- A new vignette on benchmarking was added which tests the package speed against other similar packages.
+- A new data set, `runtimes`, was included, which is exported from the colab notebook used for benchmarking here: https://colab.research.google.com/drive/1vYlBdJd4xCV43UwJ33XXpO3Ys8xWkuxx?usp=sharing
+- Sobol draws are supported via a new `drawType` argument.
+- A warning is displayed against using Halton draws after 5 random variables have been specified in a mixed logit model. Users are encouraged to switch to using Sobol draws and increasing the number of draws to at least 200.
+- Changed the argument name `price` to `scalePar` to be more general.
+- Changed the argument name `randPrice` to `randScale` to be more general.
+- The `modelSpace` argument is no longer required for specifying a WTP space model as it is redundant. Including a `scalePar` argument is enough to determine that it is a WTP space model.
+
 # logitr 0.6.1
 
 - Fixes error on some of the gradient tests from 0.6.0 release. The gradient tests were sensitive to the random starting parameters, so I modified how the starting parameters were set for the first iteration of a multistart loop such that they would be more consistent.

README.md

@@ -66,14 +66,6 @@ View the [basic
 usage](https://jhelvy.github.io/logitr/articles/basic_usage.html) page
 for details on how to use **logitr** to estimate models.
 
-## Related software
-
-If you are a Python user, [`xlogit`](https://github.com/arteagac/xlogit)
-is a similar package built in Python. It has a similar user interface
-for defining models, and it is even faster than logitr as it uses
-GPU-accelerated estimation for mixed logit models. It is a good resource
-for comparing results with those from logitr.
-
 ## Author, Version, and License Information
 
 -   Author: *John Paul Helveston* <https://www.jhelvy.com/>
@@ -94,15 +86,15 @@ citation("logitr")
 #> 
 #>   John Paul Helveston (2022). logitr: Fast Estimation of Multinomial
 #>   and Mixed Logit Models with Preference Space and Willingness to Pay
-#>   Space Utility Parameterizations. R package version 0.6.1
+#>   Space Utility Parameterizations.
 #> 
 #> A BibTeX entry for LaTeX users is
 #> 
 #>   @Manual{,
 #>     title = {logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness to Pay Space Utility Parameterizations},
 #>     author = {John Paul Helveston},
 #>     year = {2022},
-#>     note = {R package version 0.6.1},
+#>     note = {R package},
 #>     url = {https://jhelvy.github.io/logitr/},
 #>   }
 ```

cran-comments.md

@@ -7,4 +7,5 @@
 0 errors | 0 warnings | 0 notes
 
 ## Notes
-* This is a new release
+* This is a new release.
+* Apologies for another release so soon (it's been a productive week) - this will be the last for at least a month or longer.

inst/CITATION

@@ -5,7 +5,7 @@ citEntry(
   title    = "logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness to Pay Space Utility Parameterizations",
   author   = "John Paul Helveston",
   year     = "2022",
-  note     = "R package version 0.6.1",
+  note     = "R package",
   url      = "https://jhelvy.github.io/logitr/",
-  textVersion = "John Paul Helveston (2022). logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness to Pay Space Utility Parameterizations. R package version 0.6.1"
+  textVersion = "John Paul Helveston (2022). logitr: Fast Estimation of Multinomial and Mixed Logit Models with Preference Space and Willingness to Pay Space Utility Parameterizations."
 )