Stoaty said:
edatoakrun said:
Where are the "no correlation" results graphed, for the same group of bar loss reports??
I may not have posted them for the exact same data set. Here is the graph for one bar losers in Phoenix. Correlation coefficient is .0003. I think we can safely say there is no correlation.
Having no expertise in statistical analysis, I am quite willing to be corrected.
But I'd say that you could find a significant correlation in the ~50% of the reports following the trend line, and, as you say, much less correlation in the 100% of the reports graphed.
It would be very interesting if it were possible to isolate the most significant factors in common, for the ~half the LEAFs that are outliers from the (more highly correlated, 50% of reports) trend.
My best guess for the largest single "off the chart" factor, would be ambient temperature, where the LEAFs are parked at night. This probably varies significantly more, than ambient temperature during the day.