plot time series with constraint

Question

I have a big dataset and I'd like to plot zizi vs hour but only for each hour while the variables looks like this:

> datasetjc$hour[1:100]
  [1] 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23 23
 [40] 23 23 23 23 23 23 23 23 23 23 23 23 23 23  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0
 [79]  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0 

> datasetjc$zizi[1:100]
  [1]  2 27  2  3 45  0  6  0 15  8  3  1  4  0  0 15  1 13  0 15 23  8 21  2  0  9 43 26 31 33 11  0  4  7 26  2 25 14  1
 [40]  3  1  6  3  4  3  2 27  2  3 45  0  7  0 15  8  3  1  4  0  4 26  0 15  1  4  0 15 14 12 23  8  3 21 13  2  0 32 43
 [79] 31 11  4  0  4  7 26 10  2 25 25  1  1  4  4 23  3  2 27  2 45  0
>

I also have the minutes, dates and days vaiables. Each data are only separated by 5 minutes. How can I do the plot?

Thx

What do you mean "but only for each hour"? Do you mean only plot $hour data that are not `0`? I.e. all the `23`s? — WillardSolutions, Apr 21 '16 at 21:19
Looks like you have multiple entries of `zizi` for the same hour -- what do you want to display? The mean of the `zizi` values for that hour? — user295691, Apr 21 '16 at 21:20
I mean for each day, I want to plot zizi vs hour for hour equal 0, 1, 2, ... 23 while here I have steps of 5 min — BillyLeZeurbé, Apr 21 '16 at 21:46
I'm interested to know how to compute the mean of the zizi values for each hour — BillyLeZeurbé, Apr 21 '16 at 21:47

score 0 · Answer 1 · edited May 23 '17 at 12:23

If we take your question to be "how do I get hourly summaries of data taken at five minute intervals", then this is a classic split-apply-combine. Average data by group is a great summary of the different techniques.

For this particular example, in vanilla R you use the aggregate or by function.

> df <- data.frame(hour=c(1,1,2,2,3,3,4,4), zizi=1:8)
> aggregate(zizi ~ hour, data=df, mean)
  hour zizi
1    1  1.5
2    2  3.5
3    3  5.5
4    4  7.5

If you wish to summarize by date/hour, then you use +:

> aggregate(zizi ~ hour + day, data=df, mean)

For more advanced versions of this, I would recommend investing some time in learning either dplyr or data.table, as both are excellent libraries for doing more complex versions of this extremely common task.

Also for future reference, see How to make a great R reproducible example? for suggestions on how to pose a question more clearly.

plot time series with constraint

1 Answers1