Ask HN: Time series analysis

7 pointsdageroth2 comments

I am currently interested in time series analysis as it seems to be a great feature for all kinds of software to report unexcpected events. My first go is rather simple with just taking the last 90 datapoints and then look whether it differs more than 2 or 3 times from the standarddeviation. Does anyone have a simple approach for more sophisticated time series analysis? or rather, links and book recommendations?

View on HN

R has a lot of packages and functions for time series analysis. These links might get you started:

http://cran.r-project.org/web/views/TimeSeries.html

http://www.stat.pitt.edu/stoffer/tsa2/R_time_series_quick_fi...

http://cran.r-project.org/doc/contrib/Ricci-refcard-ts.pdf

http://people.bath.ac.uk/masgs/time%20series/TimeSeriesR2004...

http://www.statoek.wiso.uni-goettingen.de/veranstaltungen/ze...

http://www.r-bloggers.com/tag/time-series-charts/

The classic work on Time Series is Box-Jenkins. I've got the 3rd edition, but it looks like the 4th is now available-- I suppose I'll have to upgrade.

http://www.amazon.com/Time-Analysis-Forecasting-Probability-...

Reply on news.ycombinator.com