So Mark, which of your three different averages should one use as the
average amount a longitudinally corrected is "off" (either positive or
negative) on any randomly selected day of the year.  Is the correct answer
your mean absolute error of 7.2 minutes?  Is this what I should tell my
sundial customers?

Thanks for answering,
John Carmichael



 >John Carmichael asked:
>
>> Is the average EOT correction over the course of the entire year equal to
>> about seven or eight minutes? from (16+15)/2/2=7.75 min.. As 16 minutes is
>> about the extreme correction on Nov. 1st. and 15 minutes is the other
>> extreme corroction in mid Febuary.  Does anybody know the exact average?  
>
>
>I reported on several different error statistics for EoT in a paper I gave
>at the Seattle 1998 NASS shindig.  And I've promised Fred Sawyer that a
>write-up of my talk will eventually be submitted to _The Compendium_.
>Oh, for a few more hours in the day....
>
>
>Mean error:  0.0 minutes
>(The average error over one year, taking into account the sign of EoT.)
>
>Mean absolute error:  7.2 minutes
>(The average error when taking all EoT errors to be nonnegative.)
>
>Root-mean-square error:  8.7 minutes
>(The square root of the average of the squares of all EoT errors.)
>
>
>~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~  ~ 
>  Mark Gingrich      [EMAIL PROTECTED]      San Leandro, California
>
>

Reply via email to