1

I found three websites which lists the times of sunrises and sunsets in the place where I live. But as those times differ occasionally some minutes, I would like to know if there is a method to find out which one is the most correct prediction. I heard that statistics provides some methods to compare models but on the other hand, I guess I have to measure the goodness of times to some formula before I can say how much the error is.

So, if someone asks for a statistician how to measure which model predicts most accurately sunset and sunrise times, what kind of test he or she would use to measure the goodness?

guest
  • 73
  • 2
    Are you able to measure or find the actual sunrise/sunset times with some degree of accuracy after they have occurred to test the predictions against some known/true values? Also, is there a general agreed upon definition of sunset/sunrise that would disallow disagreeing estimates? I'm not an expert in the field, but I would expect that the estimates are based off a mathematical model with little variance other than your physical viewing location. – Underminer Feb 17 '14 at 18:23
  • I haven't found any database where are listed times that have occurred in the history. I'm sure there are formulae to estimate the times in astronomy books but I haven't any of those books. I don't know how physicists define the sunset/rise times or are there many definitions. – guest Feb 17 '14 at 18:27
  • 2
    What I'm getting at is that all the estimates could be correct. Perhaps they are estimating different things. For instance, perhaps Method 1 is accurately estimating the sunset/sunrise time at location X. Method 2 is accurately estimating the sunset/sunrise time at nearby location Y, a few miles away. Without something to compare the estimates to, it is difficult to say which is more accurate, since you do not have a target to aim at. – Underminer Feb 17 '14 at 18:52

0 Answers0