Dear R-Users,

my dataset contains timeseries that are structured into task-specific
epochs as labelled by a factor and a boolean variable for labelling each
data-point within the series as artefact or non-artefact.

Now there's no problem in addressing continuous task-specific
time-series that are free from artefacts as long as artefacts extend to
either boundary of adjacent timeseries but how can I individually select
e.g. two time-series within each task, if valid, i.e. artefact-free data
is split by artefacts that are constricted to the middle of a task?

let me give an example;
selecting the specific non-artefact timeseries works well here:

        data    task    artefact
1       ...     ...     ...
2       6046.9  task01  FALSE
3       6108.4  task02  FALSE
4       6123.3  task02  FALSE
5       6091.1  task02  FALSE
6       6118.4  task02  FALSE
7       6086.1  task02  FALSE
8       6090.5  task02  TRUE
9       6067.5  task02  TRUE
10      6215.9  task02  TRUE
11      6128.8  task02  TRUE
12      6136.7  task02  TRUE
13      6052.3  task03  TRUE
14      ...     ...     ...

in which case selecting data by the criterion of task=="task02" &
artefact==FALSE would yield a valid continuous selection ranging from
line 3 to line 7 whereas in:

        data    task    artefact
1       ...     ...     ...
2       6046.9  task01  FALSE
3       6108.4  task02  FALSE
4       6123.3  task02  FALSE
5       6091.1  task02  TRUE
6       6118.4  task02  TRUE
7       6086.1  task02  TRUE
8       6090.5  task02  TRUE
9       6067.5  task02  TRUE
10      6215.9  task02  FALSE
11      6128.8  task02  FALSE
12      6136.7  task02  FALSE
13      6052.3  task03  FALSE
14      ...     ...     ...

the same command would select data from lines 3 to 4 and then catenate
those values with data from lines 10 to 12, producing unspecific signal
change between lines 4 and 10.

Would any of you see a straightforward approach to individually select
the artefact-free epochs from the second example?

Help would be greatly appreciated!!!

Kind regards,

Alexander Holik

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to