How to thin out data in zoo

Question

Is there a function that let me thin out my data (time-series as zoo)?

The Example would be

mytime<-as.POSIXct(paste("2013-07-09 12:", c(1:59), sep=""))
mydata<-EuStockMarkets[1:59]
myts<-zoo(mydata, mytime)

I search a function that returns an irregular timeseries with just those data, that are different to the previous timestep by lets say ten points.

Any idea with zoo??

Thank you

Bastian

Answer 1

Try subset:

> subset(myts, diff(myts) > 10)
2013-07-09 12:04:00 2013-07-09 12:07:00 2013-07-09 12:10:00 2013-07-09 12:22:00 
            1621.04             1630.75             1645.89             1616.67 
2013-07-09 12:28:00 2013-07-09 12:32:00 2013-07-09 12:37:00 2013-07-09 12:38:00 
            1631.87             1650.43             1524.28             1603.65 
2013-07-09 12:39:00 2013-07-09 12:40:00 2013-07-09 12:41:00 
            1622.49             1636.68             1652.10 

Use abs(diff(myts)) if either direction is needed.