Use ?loess instead. -- Bert
Bert Gunter Genentech Nonclinical Biostatistics (650) 467-7374 "Data is not information. Information is not knowledge. And knowledge is certainly not wisdom." Clifford Stoll On Sat, Oct 4, 2014 at 12:09 AM, Grace Shi <1104271...@qq.com> wrote: > > I have to do roll regression based on the Daily data. I use the past three > weeks of daily returns as the estimation window and the regression is > estimated rolling forward one week at a time generating time series > estimates of beta. I know I should use the rollapply in zoo package. but I > am not sure how to do. > > data example: > > stock day week y x > "00001" 2009-01-02 2009-01 0.89 2.45 > "00001" 2009-01-03 2009-01 1.21 1.90 > "00001" 2009-01-04 2009-01 0.12 0.89 > "00001" 2009-01-05 2009-01 1.45 2.78 > "00001" 2009-01-06 2009-01 1.98 0.98 > "00001" 2009-01-09 2009-02 3.34 1.23 > "00001" 2009-01-10 2009-02 0.12 0.89 > "00001" 2009-01-11 2009-02 1.45 2.78 > "00001" 2009-01-13 2009-02 1.98 0.98 > "00001" 2009-01-16 2009-03 3.38 0.93 > "00001" 2009-01-17 2009-03 6.56 3.90 > "00001" 2009-01-18 2009-03 5.09 3.45 > "00001" 2009-01-19 2009-03 5.89 3.78 > > > > > [[alternative HTML version deleted]] > > ______________________________________________ > 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. ______________________________________________ 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.