R ddate 格式“%Y-%V”的问题
Issue with R ddate format "%Y-%V"
我不理解以下行为：
format(as.Date("2017-01-01"), "%Y-%V")
[1] "2017-52"
> as.Date("2017-01-01")
[1] "2017-01-01"
> format(as.Date("2017-01-01")-1, "%Y-%V")
[1] "2016-52"
作为第一行的输出，我期待得到 2016-52 而不是 2017-52，因为 1 月 1 日在 2017 年不到 4 天的一周内。
知道哪里出了问题吗？
> sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows Server >= 2012 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252
attached base packages:
[1] parallel splines stats graphics grDevices utils datasets methods base
other attached packages:
[1] plotly_4.7.1 highcharter_0.5.0 flexdashboard_0.5 xts_0.10-0 zoo_1.8-0
[6] rmdformats_0.3.3 dygraphs_1.1.1.4 pROC_1.10.0 elasticnet_1.1 lars_1.2
[11] here_0.1 brew_1.0-6 Hmisc_4.0-3 Formula_1.2-2 stringr_1.2.0
[16] devtools_1.13.3 RODBC_1.3-15 caret_6.0-77 testthat_1.0.2 rmarkdown_1.6
[21] gbm_2.1.3 lattice_0.20-35 survival_2.41-3 data.table_1.10.4 rgdal_1.2-11
[26] sp_1.2-5 colorRamps_2.3 gridExtra_2.3 DT_0.2 knitr_1.17
[31] reshape2_1.4.2 scales_0.5.0.9000 ggmap_2.6.1 dplyr_0.7.3 purrr_0.2.3
[36] readr_1.1.1 tidyr_0.7.1 tibble_1.3.4 ggplot2_2.2.1.9000 tidyverse_1.1.1
[41] pacman_0.4.6
loaded via a namespace (and not attached):
[1] readxl_1.0.0 backports_1.1.0 plyr_1.8.4 igraph_1.1.2
[5] lazyeval_0.2.1 digest_0.6.12 foreach_1.4.3 htmltools_0.3.6
[9] magrittr_1.5 checkmate_1.8.3 memoise_1.1.0 cluster_2.0.6
[13] recipes_0.1.0 modelr_0.1.1 gower_0.1.2 dimRed_0.1.0
[17] jpeg_0.1-8 colorspace_1.3-2 rvest_0.3.2 haven_1.1.0
[21] crayon_1.3.4 jsonlite_1.5 bindr_0.1 iterators_1.0.8
[25] glue_1.1.1 DRR_0.0.2 gtable_0.2.0 ipred_0.9-6
[29] questionr_0.6.1 kernlab_0.9-25 ddalpha_1.2.1 quantmod_0.4-11
[33] DEoptimR_1.0-8 maps_3.2.0 miniUI_0.1.1 Rcpp_0.12.14
[37] viridisLite_0.2.0 xtable_1.8-2 htmlTable_1.9 foreign_0.8-69
[41] mapproj_1.2-5 stats4_3.4.1 lava_1.5 prodlim_1.6.1
[45] htmlwidgets_0.9 httr_1.3.1 RColorBrewer_1.1-2 geosphere_1.5-5
[49] acepack_1.4.1 pkgconfig_2.0.1 nnet_7.3-12 rlang_0.1.2
[53] munsell_0.4.3 cellranger_1.1.0 tools_3.4.1 broom_0.4.2
[57] evaluate_0.10.1 yaml_2.1.14 ModelMetrics_1.1.0 robustbase_0.92-7
[61] RgoogleMaps_1.4.1 bindrcpp_0.2 nlme_3.1-131 mime_0.5
[65] RcppRoll_0.2.2 xml2_1.1.1 compiler_3.4.1 rstudioapi_0.7
[69] curl_2.8.1 png_0.1-7 stringi_1.1.5 highr_0.6
[73] forcats_0.2.0 Matrix_1.2-10 psych_1.7.8 httpuv_1.3.5
[77] R6_2.2.2 latticeExtra_0.6-28 bookdown_0.5 codetools_0.2-15
[81] MASS_7.3-47 assertthat_0.2.0 CVST_0.2-1 proto_1.0.0
[85] rprojroot_1.2 rjson_0.2.15 withr_2.1.0.9000 mnormt_1.5-5
[89] rlist_0.4.6.1 hms_0.3 grid_3.4.1 rpart_4.1-11
[93] timeDate_3012.100 class_7.3-14 TTR_0.23-2 shiny_1.0.5
[97] lubridate_1.6.0 base64enc_0.1-3
你的机器没有问题，这就是它的运行方式
> format(as.Date("2018-01-01"), "%Y-%V")
[1] "2018-01"
> format(as.Date("2017-01-01"), "%Y-%V")
[1] "2017-52"
%V
If the week (starting on Monday) containing 1 January has four or more
days in the new year, then it is considered week 1. Otherwise, it is
the last week of the previous year, and the next week is week 1.
(Accepted but ignored on input.)Week of the year as decimal number (01–53) as defined in ISO 8601.
第 52 周结果的解释：
来自维基百科ISO 8601
If 1 January is on a Monday, Tuesday, Wednesday or Thursday, it is in
week 01. If 1 January is on a Friday, Saturday or Sunday, it is in
week 52 or 53 of the previous year (there is no week 00). 28 December
is always in the last week of its year.
为什么是同一年？
因为它是 Week-based-year 而不是我们的日历年。
来自 Wikiedpia 的更多示例：
ISO 周编号年从第 01 周的第一天（星期一）开始，到新 ISO 年之前的星期日结束（因此没有重叠或间隙）。它包括 52 或 53 个完整的星期。一年中的第一个 ISO 周可能最多有三天实际上是在即将结束的公历年中；如果三个，则为星期一、星期二和星期三。同样，一年的最后一个 ISO 周最多可能有三天实际上是在开始的公历年中；如果三个，则为星期五、星期六和星期日。每个 ISO 周的星期四始终是公历年，由 ISO 周编号年表示。
示例：
Monday 29 December 2008 is written "2009-W01-1"
Sunday 3 January 2010 is written "2009-W53-7"
tl:博士；这是 ISO 8601 的行为，与 R 无关。为了克服这个问题，最好使用 %U
> format(as.Date("2017-01-01"), "%Y-%U")
[1] "2017-01"
> format(as.Date("2016-01-01"), "%Y-%U")
[1] "2016-00"
> format(as.Date("2016-01-02"), "%Y-%U")
[1] "2016-00"
> format(as.Date("2016-12-02"), "%Y-%U")
[1] "2016-48"
> format(as.Date("2016-12-31"), "%Y-%U")
[1] "2016-52"
我不理解以下行为：
format(as.Date("2017-01-01"), "%Y-%V")
[1] "2017-52"
> as.Date("2017-01-01")
[1] "2017-01-01"
> format(as.Date("2017-01-01")-1, "%Y-%V")
[1] "2016-52"
作为第一行的输出，我期待得到 2016-52 而不是 2017-52，因为 1 月 1 日在 2017 年不到 4 天的一周内。
知道哪里出了问题吗？
> sessionInfo()
R version 3.4.1 (2017-06-30)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows Server >= 2012 x64 (build 9200)
Matrix products: default
locale:
[1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252
[3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C
[5] LC_TIME=English_United States.1252
attached base packages:
[1] parallel splines stats graphics grDevices utils datasets methods base
other attached packages:
[1] plotly_4.7.1 highcharter_0.5.0 flexdashboard_0.5 xts_0.10-0 zoo_1.8-0
[6] rmdformats_0.3.3 dygraphs_1.1.1.4 pROC_1.10.0 elasticnet_1.1 lars_1.2
[11] here_0.1 brew_1.0-6 Hmisc_4.0-3 Formula_1.2-2 stringr_1.2.0
[16] devtools_1.13.3 RODBC_1.3-15 caret_6.0-77 testthat_1.0.2 rmarkdown_1.6
[21] gbm_2.1.3 lattice_0.20-35 survival_2.41-3 data.table_1.10.4 rgdal_1.2-11
[26] sp_1.2-5 colorRamps_2.3 gridExtra_2.3 DT_0.2 knitr_1.17
[31] reshape2_1.4.2 scales_0.5.0.9000 ggmap_2.6.1 dplyr_0.7.3 purrr_0.2.3
[36] readr_1.1.1 tidyr_0.7.1 tibble_1.3.4 ggplot2_2.2.1.9000 tidyverse_1.1.1
[41] pacman_0.4.6
loaded via a namespace (and not attached):
[1] readxl_1.0.0 backports_1.1.0 plyr_1.8.4 igraph_1.1.2
[5] lazyeval_0.2.1 digest_0.6.12 foreach_1.4.3 htmltools_0.3.6
[9] magrittr_1.5 checkmate_1.8.3 memoise_1.1.0 cluster_2.0.6
[13] recipes_0.1.0 modelr_0.1.1 gower_0.1.2 dimRed_0.1.0
[17] jpeg_0.1-8 colorspace_1.3-2 rvest_0.3.2 haven_1.1.0
[21] crayon_1.3.4 jsonlite_1.5 bindr_0.1 iterators_1.0.8
[25] glue_1.1.1 DRR_0.0.2 gtable_0.2.0 ipred_0.9-6
[29] questionr_0.6.1 kernlab_0.9-25 ddalpha_1.2.1 quantmod_0.4-11
[33] DEoptimR_1.0-8 maps_3.2.0 miniUI_0.1.1 Rcpp_0.12.14
[37] viridisLite_0.2.0 xtable_1.8-2 htmlTable_1.9 foreign_0.8-69
[41] mapproj_1.2-5 stats4_3.4.1 lava_1.5 prodlim_1.6.1
[45] htmlwidgets_0.9 httr_1.3.1 RColorBrewer_1.1-2 geosphere_1.5-5
[49] acepack_1.4.1 pkgconfig_2.0.1 nnet_7.3-12 rlang_0.1.2
[53] munsell_0.4.3 cellranger_1.1.0 tools_3.4.1 broom_0.4.2
[57] evaluate_0.10.1 yaml_2.1.14 ModelMetrics_1.1.0 robustbase_0.92-7
[61] RgoogleMaps_1.4.1 bindrcpp_0.2 nlme_3.1-131 mime_0.5
[65] RcppRoll_0.2.2 xml2_1.1.1 compiler_3.4.1 rstudioapi_0.7
[69] curl_2.8.1 png_0.1-7 stringi_1.1.5 highr_0.6
[73] forcats_0.2.0 Matrix_1.2-10 psych_1.7.8 httpuv_1.3.5
[77] R6_2.2.2 latticeExtra_0.6-28 bookdown_0.5 codetools_0.2-15
[81] MASS_7.3-47 assertthat_0.2.0 CVST_0.2-1 proto_1.0.0
[85] rprojroot_1.2 rjson_0.2.15 withr_2.1.0.9000 mnormt_1.5-5
[89] rlist_0.4.6.1 hms_0.3 grid_3.4.1 rpart_4.1-11
[93] timeDate_3012.100 class_7.3-14 TTR_0.23-2 shiny_1.0.5
[97] lubridate_1.6.0 base64enc_0.1-3
你的机器没有问题，这就是它的运行方式
> format(as.Date("2018-01-01"), "%Y-%V")
[1] "2018-01"
> format(as.Date("2017-01-01"), "%Y-%V")
[1] "2017-52"
%V
If the week (starting on Monday) containing 1 January has four or more days in the new year, then it is considered week 1. Otherwise, it is the last week of the previous year, and the next week is week 1. (Accepted but ignored on input.)Week of the year as decimal number (01–53) as defined in ISO 8601.
第 52 周结果的解释：
来自维基百科ISO 8601
If 1 January is on a Monday, Tuesday, Wednesday or Thursday, it is in week 01. If 1 January is on a Friday, Saturday or Sunday, it is in week 52 or 53 of the previous year (there is no week 00). 28 December is always in the last week of its year.
为什么是同一年？
因为它是 Week-based-year 而不是我们的日历年。
来自 Wikiedpia 的更多示例：
ISO 周编号年从第 01 周的第一天（星期一）开始，到新 ISO 年之前的星期日结束（因此没有重叠或间隙）。它包括 52 或 53 个完整的星期。一年中的第一个 ISO 周可能最多有三天实际上是在即将结束的公历年中；如果三个，则为星期一、星期二和星期三。同样，一年的最后一个 ISO 周最多可能有三天实际上是在开始的公历年中；如果三个，则为星期五、星期六和星期日。每个 ISO 周的星期四始终是公历年，由 ISO 周编号年表示。
示例：
Monday 29 December 2008 is written "2009-W01-1"
Sunday 3 January 2010 is written "2009-W53-7"
tl:博士；这是 ISO 8601 的行为，与 R 无关。为了克服这个问题，最好使用 %U
> format(as.Date("2017-01-01"), "%Y-%U")
[1] "2017-01"
> format(as.Date("2016-01-01"), "%Y-%U")
[1] "2016-00"
> format(as.Date("2016-01-02"), "%Y-%U")
[1] "2016-00"
> format(as.Date("2016-12-02"), "%Y-%U")
[1] "2016-48"
> format(as.Date("2016-12-31"), "%Y-%U")
[1] "2016-52"