-
Notifications
You must be signed in to change notification settings - Fork 2
/
NEWS
59 lines (31 loc) · 1.85 KB
/
NEWS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
Changes in cvTools version 0.3.3
+ Added functionality for cross-validation with grouped observations.
+ Bugfix in print() method for class "cvFolds": no longer produces warning.
+ Removed some examples to cut down example check times. The respective
full examples from the previous versions are available in folder
/inst/doc/examples-full as R scripts.
Changes in cvTools version 0.3.2
+ Bugfix in cvTool(): matrix of estimated prediction errors is now returned
correctly for more than one replication (this bug was introduced in the
bugfix for one replication from version 0.3.1).
+ Added tests for cvTool(), cvFit() and cvTuning() using package 'testthat'.
Changes in cvTools version 0.3.1
+ Bugfix in cvTool(): matrix of estimated prediction errors is now returned
correctly for one replication.
+ Bugfix in cvTuning(): "function" method now correctly stores matched call.
Changes in cvTools version 0.3.0
+ Changed some argument and component names regarding standard errors.
Changes in cvTools version 0.2.0
+ New generic function repCV() for (repeated) K-fold cross-validation. The
example functions cvLm(), cvLmrob() and cvLts() are now aliases for the
respective methods.
+ Bugfix in cvFit(): default names are retained if cvTool() returns a
1x1 matrix.
+ Bugfix in plot.cvSelect(): default handling of argument 'method' no
longer throws error in case of only one replication.
+ Added functionality to incorporate standard errors.
+ cvSelect() and cvTuning() now allow to specify criterion for selecting
the best model via the argument 'selectBest'.
+ New xyplot() and dotplot() methods for class "cv".
Changes in cvTools version 0.1.1
+ Added different types of obtaining CV folds.