-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathCHANGES
executable file
·200 lines (134 loc) · 7.47 KB
/
CHANGES
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
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
1.3-2 Fixed a bug in spline.correlog permutation code, minor change to plot.Sncf when add=TRUE. Minor update for compliance to new r-projects documentation standards.
1.2-9: Spell-checked and devtools / R-hub compliance
1.2-8: Updated plot.lisa file. Fixed problem with xlim argument in plot.spline.correlog and plot.Sncf. Fixed error in treatment of 'quiet' argument
1.2-7: Fixed error in plot.lisa help file.
1.2-6: Fixed stupid coding mistake in print.mSynch
1.2-5: Final tune-up for complete compliance with github, Rstudio and devtools
1.2-4: Fixed ... in plot-function
1.2-3: Rewrote source to use devtools/roxygen (thanks to Jun Cai). Fixed a silly error in summary.Sncf. Changed name of spline.correlog.2D to spline.correlog2D for consistency (hope this wont cause trouble). Linked source to github.
1.2-2: DOIs added. Fixed calculations with latlon data.
1.2-1: New plot.cc.offset-function. Fixed a stupid double loop (and vectorized gcdist).
Vectorized spherical correlation function (added circ.cor2) to greatly speed calculations.
Prettied up the progress counters and turn default for quiet back to FALSE. Major under-the-hood re-code to ease maintenance.
1.1-8: Plotting functions for Spline.correlogs and Sncf changed
Default for 'quiet' switched to TRUE
print.Sncf and print.spline.correlog added.
Bit of code tidying and CRAN compatibility fixing.
1.1-7: Small changes to plot.lisa.nc, plot.lisa and spatial.plot.
Bugs in correlog and correlog.nc when latlon=TRUE fixed
1.1-6: Added a new function to calculate mantel.correlog() rams,
updated lisa and lisa.nc, added a new spatial.plot() function.
Compliant with new CRAN checks
1.1-5: Some minor changes to three plotting functions
1.1-4: The auxcilliary function cor2() was changed
1.1-3: error in partial.mantel.test corrected
1.1-2: Sncf tweaked so that the cross-covariance function will use the
predicted y-intercept if z and w are sampled at different locations.
Otherwise the mean of the diagonal of the cross-correlation matrix
is used (as before).
quiet argument added to all functions to surpress bootstrap counter
(if desired)
1.1-1: R-2.6 compliant, cran compliant
1.1-0: Added lisa, added fixed bug in correlog
(when resamp=0). nc.correlog and nc.lisa renamed correolg.nc and lisa.nc,
and appropriate permutation tests implemented.
Various minor fixes in documentation
1.0-9: Calculation of distance classes corrected in
correlog and nc.correlog to make it consistent with spdep
(thanks to Ingolf Kuhn)
Correct some grammatical errors in the documentation.
1.0-8: A function for partial Mantel test added
1.0-7: Jitter argument included in Sncf.srf
1.0-6: Documentation w.r.t. latlong argument corrected
(thanks to Amit Vora)
1.0-5: A 'circ' argument was added to Sncf and mSynch functions
to allow calculations with angular data (in radians);
cor2 function updated for 1.9
Some inconsequential garbage in the code was cleaned up.
Compliant with R 1.9
Deleted some old obsolete functions (oldSncf2D, inner.na)
1.0-4: Two functions (spline.correlog.2d, plot.cc.offset)
and one dataset (lbm) added
1.0-3: cor2 auxilliary function changed in response to R-1.8
changes in cor-syntax
1.0-2: FF-filtering fixed (thanks to Ben Bolker)
1.0-0: First propper R packaging
Previous Version History:
June 16, 2003: spline.correlog enhanced to calculate
e-intercepts
June 16, 2003: rmvn.spa updated to take an optional nugget.
April 26, 2003: spline.correlog.2d the unvariate analogue to m2d
currently in beta testing.
October 11, 2002: m2d an alternative to Sncf2d incorporated in
core source.
March 21, 2002: A couple of R/S+ inconsistencies corrected.
(atan of S+ subsumes atan2 of R). library(modreg) is
automatically called under R.
March 17, 2002: A number of problems corrected to ensure
S4 compliance. (functions now run under S-plus 6,
and R 1.4)
March 17, 2002: x-intercept calculation updated in all
functions to ensure more meaningful calculations
under cross-correlation (and under other cases of
locally negative correlation) (as suggested by Eric
Seabloom)
March 17, 2002: a permutation test for significance added
to correlog. A new function mantel.test for a
permutation based mantel test added [currently in beta]
March 17, 2002: gcdist and inner.na functions written as separate
functions.
October 10, 2001: Install page rewritten
September 20, 2001: Error in calculations of cross-correlation
corrected across all functions. (Thanks to Patrick Tobin).
September 2, 2001: Incorporated code of Tim Keitt for numerically
improved algorithm for calculating spatial distances for
latitude/longitude data
September 1, 2001: error fixed in calculation of cross-correlation
function for Sncf.srf.
August 12, 2001: mSynch have the added ability to calculate
cross-correlation
August 5-13, 2001: Sncf2D made R-compliant.
July 10-13, 2001: Recoded to be R compliant. All functions (except
Sncf2D) are fully R compliant!
May-June, 2001: All functions have the added ability to calculate
cross-correlation, cross-correlograms, cross-covariances, etc.
April 24, 2001: corrected bug in sncf.q file
March 22, 2001: nc.lisa added. rmvn.spa added.
March 21, 2001: Reformating and recompilation of the help-files.
October 08, 2000: An option 'latlon' was added to Sncf allowing
geographic coordinates to be given in latitude and longitude.
The correct distance matrix (in km) is calculated.
August 16, 2000: Sncf.na consolidated with Sncf; Sncf2D.na consolidated
with Sncf2D. Code branches to pairwise deletion of missing
values (if na.omit = T)
August 16, 2000: coded mSync to calculate the mean regional correlation
(with a bootstrap CI) from a spatiotemporal data-panel.
July 17, 2000: Sncf2d (and Sncf2D.na) for the anisotropic correlation
functions have been included in the core distribution. Sncf.cc
to calculate the cross-correlation functions of sympatric
data-panels. ncf (and related) are, now, deleted from the
distribution.
June 13, 2000: rewrite and recode with perticular reference to html
help files. Sncf, Sncf.srf, and Sncf.na is now distributed as
part of the core library.
April 19, 2000: coded Sncf.srf which is like Sncf, but for spatiotemporal
data-panels from stationary random field (marginal expectation
and variance identical for n*p random variates). [Currently
distributed in in beta]
April 18, 2000: coded Sncf.na which is like Sncf, but handles (through
pair-wise deletion) spatiotemporal data-panels with missing
data. [Currently distributed in beta]
April 18, 2000: coded Sncf (eventually to replace ncf) which have the
added funcionality of calculating regional synchrony ('cbar'
with CI), and cbar-intercept (the distance at which synchrony
decends to the region wide average). [Currently distributed in
beta: sncf.q]
February 14, 2000: a error in the ncf-function description was corrected
January 21, 2000: An error in the newton search for the x-intercept and
e-intercept was corrected. (Thanks to Ole Chr. Lingjaere)
January 20, 2000: An error in the fourier filter was corrected.
(Thanks to Ole Chr. Lingjaere)
January 20, 2000: An error in the bootstrap algorithm was corrected.
(Thanks to Peter Thrall)
December 07, 1999: An error in the bootstrap algorithm was corrected
November 12, 1999: Initial code uploaded