There are several solutions available depending on the particular problem, some of them statistically "cleaner" than others:
R:
Is there an easy way to get the acf and pacf for an irregular times series? That is, the acf and pacf with lag lengths that are in units of time, not observation number.
For example eliminate non-business days (NA's) from the series and compute the acf and pacf (e.g. with na.remove from tseries).
For example interpolate to get a regular series and compute acf and pacf (e.g. with approx.irts from tseries).
For example use a methodology which can treat NA's (e.g. use Kalman filtering from ts (R-1.8.1) now ??) and compute the acf and pacf from the estimated model...
best Adrian
Thanks,
Jason Higbee
Research Associate
Federal Reserve Bank of St. Louis
The views expressed in this email are the author's and not necessarily those of the Federal Reserve Bank of St. Louis or the Federal Reserve System
[[alternative HTML version deleted]]
______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
