christophsax / timemachine Goto Github PK
View Code? Open in Web Editor NEWTravel Through Time
Travel Through Time
I have a dataset with 100s of time series, now I would like to generate a pseudo realtime data set with pseudo_history().
For each series I know its publication lag:
series_name pub_lag
adp employment change 3
T5YIE 0
How should I set up the command pseudo_history() for such a use case?
library(timemachine)
#> Loading required package: miniUI
library(forecast)
#> Registered S3 method overwritten by 'xts':
#> method from
#> as.zoo.xts zoo
#> Registered S3 method overwritten by 'quantmod':
#> method from
#> as.zoo.data.frame zoo
#> Registered S3 methods overwritten by 'forecast':
#> method from
#> fitted.fracdiff fracdiff
#> residuals.fracdiff fracdiff
dates = seq(
from = as.Date("2017-07-01"),
to = as.Date("2017-10-01"),
by = "month"
)
mytimemachine <- function(idname) {
idname <<- idname # global assignement needed!
timemachine(
aa = {
m <- forecast(auto.arima(get(idname)), h = 3)
m$mean
},
history = swiss_history,
dates = dates
)
}
mytimemachine("GDP.CH")
#> 2017-07-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-08-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-09-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-10-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> # A tibble: 12 x 4
#> expr pub_date ref_date value
#> <chr> <date> <date> <dbl>
#> 1 aa 2017-07-01 2017-07-01 168678.
#> 2 aa 2017-07-01 2017-10-01 169190.
#> 3 aa 2017-07-01 2018-01-01 169708.
#> 4 aa 2017-08-01 2017-07-01 168678.
#> 5 aa 2017-08-01 2017-10-01 169190.
#> 6 aa 2017-08-01 2018-01-01 169708.
#> 7 aa 2017-09-01 2017-07-01 168678.
#> 8 aa 2017-09-01 2017-10-01 169190.
#> 9 aa 2017-09-01 2018-01-01 169708.
#> 10 aa 2017-10-01 2017-10-01 170404.
#> 11 aa 2017-10-01 2018-01-01 171072.
#> 12 aa 2017-10-01 2018-04-01 171675.
mytimemachine("GC")
#> 2017-07-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-08-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-09-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> 2017-10-01
#> Attached: GDP.CH, GC, PC, INV, INV.CONSTR, INV.EQ, EXP, IMP, DEFL, CPI.NSA, CPI.SA, I, RR, EMPL.NSA, EMPL.SA, RU.NSA, RU.SA, UR.NSA, UR.SA, GDP.F, GDP.US, GDP.EA, GDP.JP
#> # A tibble: 12 x 4
#> expr pub_date ref_date value
#> <chr> <date> <date> <dbl>
#> 1 aa 2017-07-01 2017-07-01 19928.
#> 2 aa 2017-07-01 2017-10-01 19984.
#> 3 aa 2017-07-01 2018-01-01 20050.
#> 4 aa 2017-08-01 2017-07-01 19928.
#> 5 aa 2017-08-01 2017-10-01 19984.
#> 6 aa 2017-08-01 2018-01-01 20050.
#> 7 aa 2017-09-01 2017-07-01 19928.
#> 8 aa 2017-09-01 2017-10-01 19984.
#> 9 aa 2017-09-01 2018-01-01 20050.
#> 10 aa 2017-10-01 2017-10-01 20006.
#> 11 aa 2017-10-01 2018-01-01 20077.
#> 12 aa 2017-10-01 2018-04-01 20147.
Created on 2019-08-20 by the reprex package (v0.3.0)
Wie kann den Forecast Horizon h so setzen, dass es nicht die Quartale als Referenz nimmt, also von 1 bis 9, sondern die Monate, sprich von 1 bis 27? Der Code unten macht sehr schön die Quartale.
results <- structure(list(expr = c("bdg1", "bdg1", "bdg1", "bdg1", "bdg1",
"bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1",
"bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1",
"bdg1", "bdg1", "bdg1", "bdg1", "bdg1", "bdg1"), pub_date = structure(c(14214,
14245, 14276, 14304, 14335, 14365, 14396, 14426, 14457, 14488,
14518, 14549, 14579, 14610, 14641, 14669, 14700, 14730, 14761,
14791, 14822, 14853, 14883, 14914, 14944, 14975, 15006), class = "Date"),
ref_date = structure(c(14883, 14883, 14883, 14883, 14883,
14883, 14883, 14883, 14883, 14883, 14883, 14883, 14883, 14883,
14883, 14883, 14883, 14883, 14883, 14883, 14883, 14883, 14883,
14883, 14883, 14883, 14883), class = "Date"), value = c(-0.506681784942139,
-0.571015209472469, 0.142607140892838, -0.156631763687276,
-0.242282371579598, -0.129473293377144, 0.00508796161221747,
0.871150764454464, 0.86779210729246, 0.798360420519133, 0.793553987258127,
0.801069194315595, 0.792874354495371, 0.739087556088245,
0.760095102547694, 0.750505127575601, 1.10703202479419, 1.11619016768242,
1.17228081482969, 1.1391799835059, 1.24485315943325, 1.00193624512288,
1.08847534736896, 1.20249116401625, 1.04852976364214, 1.02529181687138,
1.02529181687138)), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -27L), .Names = c("expr", "pub_date", "ref_date",
"value"))
errors <-
results %>%
group_by(ref_date, pub_date, expr) %>%
mutate(h = seq(n())) %>%
ungroup()
A tibble: 27 x 4
expr pub_date ref_date value
1 bdg1 2008-12-01 2010-10-01 -0.507
2 bdg1 2009-01-01 2010-10-01 -0.571
3 bdg1 2009-02-01 2010-10-01 0.143
4 bdg1 2009-03-01 2010-10-01 -0.157
5 bdg1 2009-04-01 2010-10-01 -0.242
6 bdg1 2009-05-01 2010-10-01 -0.129
7 bdg1 2009-06-01 2010-10-01 0.00509
8 bdg1 2009-07-01 2010-10-01 0.871
9 bdg1 2009-08-01 2010-10-01 0.868
10 bdg1 2009-09-01 2010-10-01 0.798
... with 17 more rows
hier sollte es dann also bei 27 beginnen und dann runterzählen. Any help kindly appreciated!
Frage 1: wie muss ich vorgehen um eine Vielzahl von Monatsindikatoren in die pseudo_history zu bringen?
Frage 2: ist es danach korrekt, diese neu generierte Tabelle mit der pseudo_history der target variable zu mergen via ts_c() und auf dieses neue Objekt options(timemachine.history = NEW_OBJECT) anzuwenden?
library("timemachine")
library("tsbox")
dta <- structure(c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, 61.481006, 58.857049, 59.242728,
57.394994, 56.780823, 56.132335, 53.96825, 53.177727, 53.78973,
50.352191, 48.597276, 47.09809, 50.068857, 49.220064, 48.4233,
44.834267, 40.356769, 45.781857, 47.261783, 46.636696, 47.281833,
47.583723, 46.105661, 49.914601, 50.656413, 53.183586, 52.724529,
53.893057, 54.83731, 55.900363, 60.681033, 56.959559, 57.116975,
57.975628, 59.559429, 58.937327, 57.440081, 57.288777, 57.22586,
57.72176, 56.765583, 54.697867, 55.700123, 54.786896, 50.699391,
50.154701, 49.896231, 50.826276, 46.251106, 48.171144, 49.283641,
51.222098, 51.942546, 52.731403, 55.522627, 55.47006, 58.808512,
61.732733, 61.479249, 61.661505, 58.356902, 60.595626, 61.90567,
63.762693, 63.815429, 64.088621, 60.153999, 61.997755, 62.549996,
63.026796, 61.329778, 59.550411, 57.167653, 54.717035, 51.534965,
46.904027, 47.620669, 47.366377, 44.222626, 43.292673, 44.667411,
41.40868, 39.215695, 41.308652, 44.677088, 46.705239, 48.015919,
48.748946, 50.130922, 49.504175, 49.383163, 46.527406, 46.074503,
46.509469, 52.169285, 46.32808, 48.015995, 49.290532, 45.965258,
45.822409, 43.219604, 43.698029, 48.541393, 51.280682, 52.157872,
54.278471, 55.049091, 53.81716, 56.489512, 53.881038, 58.876912,
58.569519, 57.889117, 57.621017, 54.845328, 55.234341, 56.015731,
54.80292, 53.679822, 54.050851, 52.490971, 47.615934, 51.724028,
51.200589, 52.37204, 51.94741, 51.783009, 50.676276, 58.523104,
58.084198, 56.468492, 56.592854, 58.481711, 60.903702, 62.779868,
63.472528, 63.804721, 63.864678, 64.480932, 65.853775, 65.779066,
63.466652, 66.175509, 66.906063, 62.553569, 64.283554, 61.222021,
61.794463, 59.876203, 62.87484, 62.485397, 62.886291, 59.406877,
61.105834, 61.906172, 61.875533, 61.892313, 60.699477, 58.018542,
55.777674, 55.330439, 54.630483, 52.377589, 51.676317, 48.909912,
46.290755, 36.170746, 35.089496, 36.441261, 33.501931, 32.879724,
35.420966, 39.561177, 41.910829, 43.590895, 49.386207, 54.959863,
54.959504, 55.592536, 54.771514, 56.957461, 57.426285, 63.71277,
64.108108, 65.16279, 63.949701, 65.505551, 61.298826, 62.32585,
62.145334, 62.678882, 60.625537, 60.505899, 62.040462, 58.457601,
57.996903, 57.753367, 53.866712, 52.168522, 52.038234, 50.588769,
49.004712, 45.819341, 49.333427, 47.910997, 47.950096, 50.568051,
46.709538, 45.890086, 47.724321, 46.682034, 46.816451, 45.605033,
46.778456, 48.5428, 49.947659, 52.323094, 50.040567, 49.655981,
48.972675, 52.415164, 51.684275, 55.21799, 55.52062, 56.806439,
54.637304, 56.972046, 54.120359, 56.163625, 56.977715, 54.32479,
54.86243, 53.770848, 54.076539, 53.399577, 54.073174, 51.565473,
54.192213, 51.690495, 52.599805, 49.612651, 47.971448, 47.754378,
47.873692, 47.845236, 49.789611, 49.590881, 51.287462, 48.088433,
49.622844, 48.71384, 49.752088, 50.816714, 51.067484, 52.832851,
52.982981, 56.279048, 51.108466, 50.875039, 51.698094, 54.519177,
55.239914, 55.513878, 55.887999, 54.974043, 58.094389, 57.464562,
58.335638, 55.719123, 59.980663, 60.279548, 61.412128, 62.177216,
61.50319, 64.585728, 65.561556, 65.296017, 65.450934, 60.276087,
63.620851, 62.415046, 61.580441, 61.930521, 64.814788, NaN, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, 99.501305, 100.248382, 100.078951, 94.186669,
96.292492, 91.653911, 92.254047, 88.492059, 89.957219, 88.69435,
86.673418, 86.94696, 90.879503, 89.245105, 89.656445, 96.228874,
93.898861, 97.650219, 98.828576, 101.612057, 101.997869, 101.230586,
105.687884, 111.427507, 110.749815, 108.008646, 107.369113, 109.961889,
107.615364, 108.36063, 112.317991, 112.040318, 112.34915, 114.397031,
107.984066, 106.861048, 104.054958, 97.774113, 102.954049, 100.214249,
104.162704, 98.027929, 97.826268, 96.634005, 96.181744, 93.022481,
87.523834, 91.269186, 95.553449, 102.015698, 99.318494, 103.137761,
102.689395, 103.374937, 107.234646, 108.178722, 110.76927, 114.094632,
114.43496, 112.621698, 104.472523, 106.974185, 111.638104, 110.683348,
108.17878, 111.77668, 105.124857, 102.201072, 102.097924, 102.852667,
103.199146, 102.302014, 99.328182, 95.263555, 93.335938, 88.044439,
83.434603, 87.332845, 90.542983, 94.122428, 91.891692, 83.917714,
80.848436, 83.479795, 93.799296, 99.408233, 100.196253, 100.913541,
105.632358, 101.652076, 95.776375, 95.999099, 93.719823, 94.702188,
95.180696, 95.852128, 97.278011, 95.832943, 94.081202, 92.808528,
95.810806, 96.295982, 101.979694, 102.741928, 104.441418, 111.922919,
113.534041, 111.76951, 110.421103, 112.463349, 112.150355, 113.731978,
111.227388, 109.8938, 107.910234, 107.536339, 106.052374, 101.115117,
98.959026, 100.390357, 100.055504, 101.022068, 100.387432, 100.481601,
101.818311, 103.339738, 103.34459, 101.674352, 103.779091, 107.83464,
108.71202, 109.341725, 111.635017, 111.975905, 111.041064, 110.354663,
108.732337, 106.884462, 106.973457, 107.090058, 108.87131, 108.20023,
107.828296, 107.164468, 110.112013, 107.451683, 109.610297, 104.386831,
103.004782, 103.33263, 105.106871, 104.211926, 101.646796, 97.961011,
101.164349, 101.056704, 97.462079, 94.811914, 92.185058, 83.332929,
85.383903, 83.683249, 81.917405, 81.3495, 81.771502, 80.038802,
71.377957, 67.772135, 70.780808, 74.866347, 76.314171, 88.689248,
97.842247, 107.227517, 111.916082, 115.742203, 117.615547, 118.853487,
121.427066, 117.695849, 112.026158, 116.307885, 113.94133, 117.427694,
116.597738, 111.231593, 109.819555, 109.607432, 109.27713, 103.82075,
106.181738, 107.781631, 102.174211, 101.750021, 103.25056, 103.254782,
99.105551, 98.105074, 89.072696, 85.124292, 82.877088, 84.248012,
87.987261, 92.519579, 99.417577, 101.121926, 103.288251, 102.043409,
97.338767, 95.416932, 98.204944, 100.976193, 100.857232, 101.675736,
101.748545, 101.583094, 105.787254, 106.223878, 104.887749, 101.617687,
100.307647, 101.420607, 106.181387, 106.341568, 108.13514, 107.287109,
108.888897, 105.594897, 106.019028, 106.357739, 104.693886, 104.197591,
100.753477, 99.991623, 98.01718, 100.365854, 97.206255, 99.69125,
97.559906, 98.024285, 95.511774, 86.479443, 88.911985, 89.079421,
103.234515, 98.576245, 100.911628, 100.652339, 100.606693, 101.97321,
99.127473, 98.105399, 101.766192, 103.002255, 101.987832, 103.612223,
102.051776, 102.08772, 101.075767, 100.221771, 100.969933, 102.71923,
102.416446, 103.089981, 102.742355, 108.16279, 108.147366, 107.468454,
103.863916, 105.32695, 107.434802, 104.427945, 106.144063, 109.245349,
110.286943, 111.437629, 107.517935, 108.255689, 105.037057, 103.293145,
100.013446, 101.256753, 101.68433, 100.302153, NaN, NA, NA, NA
), .Dim = c(468L, 2L), .Dimnames = list(NULL, c("pmi.ind_tot",
"kofbaro")), .Tsp = c(1980, 2018.91666666667, 12), class = c("mts",
"ts", "matrix"))
# does not work
pseudo_history(dta, "1 month")
# does not work
dta %>%
ts_tbl() %>%
group_by(id) %>%
pseudo_history("1 month")
# does not yield pseudo real-time dataset
indic.pseudo_rt <- dta %>%
ts_tbl() %>%
rename(ref_date = "time") %>%
rename(var = "id") %>%
mutate(pub_date = add_to_date(ref_date, "1 month")) %>%
select(pub_date, ref_date, var, value)
Ich habe bereits Pakete deinstalliert (miniUI und timemachine), RStudio neu gestartet, die Pakets nochmal installiert, aber es funktioniert immer nicht.
√ checking for file 'C:\Users\David\AppData\Local\Temp\RtmpmwbsmJ\remotes1db465d3324\christophsax-timemachine-b5eb705/DESCRIPTION' (472ms)
- preparing 'timemachine':
√ checking DESCRIPTION meta-information ...
- checking for LF line-endings in source and make files and shell scripts
- checking for empty or unneeded directories
- building 'timemachine_0.0.3.tar.gz'
Installing package into ‘C:/Users/David/Documents/R/win-library/4.0’
(as ‘lib’ is unspecified)
* installing *source* package 'timemachine' ...
** using staged installation
** R
** data
*** moving datasets to lazyload DB
** inst
** byte-compile and prepare package for lazy loading
Error: (converted from warning) package 'miniUI' was built under R version 4.0.5
Execution halted
ERROR: lazy loading failed for package 'timemachine'
* removing 'C:/Users/David/Documents/R/win-library/4.0/timemachine'
Error: Failed to install 'timemachine' from GitHub:
(converted from warning) installation of package ‘C:/Users/David/AppData/Local/Temp/RtmpmwbsmJ/file1db42e9090c/timemachine_0.0.3.tar.gz’ had non-zero exit status
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.