thr3ads.net - R help - [R] Time series data decomposition from by minute data [Oct 2024]

If this information is useful, please help other people find it:
Share via:

roslinazairimah zakaria

2024-Oct-03 13:24 UTC

[R] Time series data decomposition from by minute data

Dear all,

My data is by minutes and I can see it has seasonal trend by daily and
weekly. How do I decompose the minute data into daily and weekly

some data:
> dput(tail(dt_train,100))structure(c(11L, 11L, 10L, 10L, 10L, 10L, 10L, 10L,
10L, 10L,10L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
10L, 11L, 11L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
10L, 10L, 10L, 10L, 10L, 10L, 10L, 9L, 9L, 9L, 8L, 8L, 8L, 8L,
8L, 8L, 8L, 8L, 8L, 7L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L,
8L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
9L, 9L, 9L), class = c("xts", "zoo"), index =
structure(c(1412622480,
1412622540, 1412622600, 1412622660, 1412622720, 1412622780, 1412622840,
1412622900, 1412622960, 1412623020, 1412623080, 1412623140, 1412623200,
1412623260, 1412623320, 1412623380, 1412623440, 1412623500, 1412623560,
1412623620, 1412623680, 1412623740, 1412623800, 1412623860, 1412623920,
1412623980, 1412624040, 1412624100, 1412624160, 1412624220, 1412624280,
1412624340, 1412624400, 1412624460, 1412624520, 1412624580, 1412624640,
1412624700, 1412624760, 1412624820, 1412624880, 1412624940, 1412625000,
1412625060, 1412625120, 1412625180, 1412625240, 1412625300, 1412625360,
1412625420, 1412625480, 1412625540, 1412625600, 1412625660, 1412625720,
1412625780, 1412625840, 1412625900, 1412625960, 1412626020, 1412626080,
1412626140, 1412626200, 1412626260, 1412626320, 1412626380, 1412626440,
1412626500, 1412626560, 1412626620, 1412626680, 1412626740, 1412626800,
1412626860, 1412626920, 1412626980, 1412627040, 1412627100, 1412627160,
1412627220, 1412627280, 1412627340, 1412627400, 1412627460, 1412627520,
1412627580, 1412627640, 1412627700, 1412627760, 1412627820, 1412627880,
1412627940, 1412628000, 1412628060, 1412628120, 1412628180, 1412628240,
1412628300, 1412628360, 1412628420), tzone = "", tclass =
c("POSIXct",
"POSIXt")), dim = c(100L, 1L))


I also attached the plot of training data.


I tried :

decompose(dt_train, type = "multiplicative", filter = NULL)Error in
decompose(dt_train, type = "multiplicative", filter = NULL) :
  time series has no or less than 2 periods

> stl(dt_train, s.window = "periodic")Error in stl(dt_train,
s.window = "periodic") :  series is not periodic or has less than two


-- 
*Roslinazairimah Zakaria*
*Tel: +609-5492370; Fax. No.+609-5492766*

*Email: roslinazairimah at ump.edu.my <roslinazairimah at ump.edu.my>;
roslinaump at gmail.com <roslinaump at gmail.com>*
Faculty of Industrial Sciences & Technology
University Malaysia Pahang
Lebuhraya Tun Razak, 26300 Gambang, Pahang, Malaysia

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Rplot_dt_train.pdf
Type: application/pdf
Size: 217119 bytes
Desc: not available
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20241003/02f10b2a/attachment.pdf>

Petr Pikal

2024-Oct-04 13:18 UTC

head link

[R] Time series data decomposition from by minute data

Hallo

you can extract POSIX object

tv <- as.POSIXct(index(dt_train))

and use cut together with aggregate
cut(tv, "hour")

aggregate(dt_train, list(cut(tv, "hour")), mean)

2014-10-06 21:00:00 9.807692
2014-10-06 22:00:00 8.666667

Cheers.
Petr



?t 3. 10. 2024 v 17:25 odes?latel roslinazairimah zakaria <
roslinaump at gmail.com> napsal:
> Dear all,
>
> My data is by minutes and I can see it has seasonal trend by daily and
> weekly. How do I decompose the minute data into daily and weekly
>
> some data:
>
> > dput(tail(dt_train,100))structure(c(11L, 11L, 10L, 10L, 10L, 10L, 10L,
> 10L, 10L, 10L,
> 10L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
> 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
> 10L, 11L, 11L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L, 10L,
> 10L, 10L, 10L, 10L, 10L, 10L, 10L, 9L, 9L, 9L, 8L, 8L, 8L, 8L,
> 8L, 8L, 8L, 8L, 8L, 7L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L, 8L,
> 8L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L, 9L,
> 9L, 9L, 9L), class = c("xts", "zoo"), index =
structure(c(1412622480,
> 1412622540, 1412622600, 1412622660, 1412622720, 1412622780, 1412622840,
> 1412622900, 1412622960, 1412623020, 1412623080, 1412623140, 1412623200,
> 1412623260, 1412623320, 1412623380, 1412623440, 1412623500, 1412623560,
> 1412623620, 1412623680, 1412623740, 1412623800, 1412623860, 1412623920,
> 1412623980, 1412624040, 1412624100, 1412624160, 1412624220, 1412624280,
> 1412624340, 1412624400, 1412624460, 1412624520, 1412624580, 1412624640,
> 1412624700, 1412624760, 1412624820, 1412624880, 1412624940, 1412625000,
> 1412625060, 1412625120, 1412625180, 1412625240, 1412625300, 1412625360,
> 1412625420, 1412625480, 1412625540, 1412625600, 1412625660, 1412625720,
> 1412625780, 1412625840, 1412625900, 1412625960, 1412626020, 1412626080,
> 1412626140, 1412626200, 1412626260, 1412626320, 1412626380, 1412626440,
> 1412626500, 1412626560, 1412626620, 1412626680, 1412626740, 1412626800,
> 1412626860, 1412626920, 1412626980, 1412627040, 1412627100, 1412627160,
> 1412627220, 1412627280, 1412627340, 1412627400, 1412627460, 1412627520,
> 1412627580, 1412627640, 1412627700, 1412627760, 1412627820, 1412627880,
> 1412627940, 1412628000, 1412628060, 1412628120, 1412628180, 1412628240,
> 1412628300, 1412628360, 1412628420), tzone = "", tclass =
c("POSIXct",
> "POSIXt")), dim = c(100L, 1L))
>
>
> I also attached the plot of training data.
>
>
> I tried :
>
> decompose(dt_train, type = "multiplicative", filter = NULL)Error
in
> decompose(dt_train, type = "multiplicative", filter = NULL) :
>   time series has no or less than 2 periods
>
>
> > stl(dt_train, s.window = "periodic")Error in stl(dt_train,
s.window > "periodic") :
>   series is not periodic or has less than two
>
>
> --
> *Roslinazairimah Zakaria*
> *Tel: +609-5492370; Fax. No.+609-5492766*
>
> *Email: roslinazairimah at ump.edu.my <roslinazairimah at
ump.edu.my>;
> roslinaump at gmail.com <roslinaump at gmail.com>*
> Faculty of Industrial Sciences & Technology
> University Malaysia Pahang
> Lebuhraya Tun Razak, 26300 Gambang, Pahang, Malaysia
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> https://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]

Possibly Parallel Threads

Search for more possibly parallel threads

R help - Oct 2024 - Time series data decomposition from by minute data

[R] Time series data decomposition from by minute data

[R] Time series data decomposition from by minute data

Possibly Parallel Threads