use log return or quotient return?
Many markets use quotient returns or discounts. Returns expressed in this way are not consistent between markets. Also aggregating over time is only approximate. For this reason empirical work would be better served if the institutional rates were converted to log difference returns before analysis. Even for macroeconometric analysis of various growth rates log differences are more aligned with theory and are likely to give better results. I would recommend that, regardless of market practices, all serious empirical work on returns be done in terms of log differences John C. Frain. Economics Department Trinity College Dublin Dublin 2 www.tcd.ie/Economics/staff/frainj/home.html mailto:frainj at tcd.ie Quoting Krishna Kumar <kriskumar at earthlink.net>:
There are atleast three ways to compute returns take first differences , take first differences and scale, take first differences of the log returns One of the nice aspect of first differences of log is that they include scaling and all the three are approximately the same(as Gabor points out) at high-freq over a short period of time. But if you had a lower freq data over a much longer period of time then it is useful to investigate the statistical properties of the returns before going one way or the other. Best, Krishna Gabor Grothendieck wrote:
It depends on how good the approximation log(1+r) = r is and that depends on whether r is sufficiently small or not. On 11/13/06, Michael <comtech.usa at gmail.com> wrote:
Hi all, Does anybody know which is more commonly used in financial time series -- log return or quotient return? Thanks a lot,
_______________________________________________ R-SIG-Finance at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance
_______________________________________________ R-SIG-Finance at stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance