Reader Comments

Post a new comment on this article

Analysis is incorrect

Posted by Liebhold on 26 Feb 2012 at 02:04 GMT

was well described by a linear equation with common time trend
http://plosone.org/article/info:doi/10.1371/journal.pone.0024733#article1.body1.sec3.p1

This analysis is incorrect because the trend should have been removed prior to testing for the effect of the covariate (i.e. temperature). The authors should have first detrended the series and THEN tested for the effect of temperature. This analysis violates a basic assumption of the analysis, namely stationarity.

No competing interests declared.