descriptionTime-series sAmpLes forecasted With ExoGenous variables
last changeThu, 6 Feb 2020 07:57:07 +0000 (08:57 +0100)

Predict PM10 as functions of time

Joint work with Jean-Michel Poggi and Bruno Portier

Forecast a curve sampled within the day (seconds, minutes, hours…), using past measured curves + past exogenous informations, which could be some aggregated measure on the past curves, the weather… Main starting point: computeForecast().

NOTE: algorithms are not specific to PM10 and could be applied to anything else (in similar contexts). However, seasons are hard-coded to follow pollution events. You may want to change them in pkg/R/utils.R, function .isSameSeason()

NOTE 2: the package works for only one series (one station). Further extensions might consider the multi-series case.

The final report may be found at this location

2020-02-06 Benjamin Auder'update' master
2020-02-06 Benjamin Audermerge with remote
2018-01-22 Benjamin AuderFirst commit
2017-07-02 Benjamin Auderfix bugs
2017-06-29 Benjamin Auderfix Data acquisition; TODO: check CV for local method
2017-06-20 Benjamin Auderupdate following 23/05 TODOs
2017-05-30 Benjamin Auderremove merging marks
2017-05-30 Benjamin Auderafter merg
2017-05-30 Benjamin Auderbetter style in report
2017-05-30 Benjamin Auderprepare last changes
2017-05-18 Benjamin Auderafter NA fixes; increase Jupyter cell timeout
2017-05-18 Benjamin Auder'update'
2017-05-17 Benjamin Auderprepare package for yearly report
2017-05-17 Benjamin AuderCLEANING
2017-05-11 Benjamin Auderprepare report on one full year (2015)
2017-05-04 Benjamin Auder'update'
3 years ago master