Quote:
Originally Posted by Citrus Dad
Since this is a linear regression should be able to produce two fundamental statistics that tell about goodness of fit. (There's other stats that also can tell us about potential bias as well, but those are more difficult in a spreadsheet.) First is to compute the standard error around each estimate so we can see the probability that the parameter estimates are statistically significant. The second is the r-squared that tells about the overall goodness of fit.
|
Since the assumption of normal distribution is almost certainly false, I think a
cumulative histogram of unpenalized score Ax-b residuals may provide better insight concerning how well the data match the OPR model.
However, if someone would be willing to extract from Ed's spreadsheet his carefully crafted unpenalized alliance scores and team composition into a simple CSV or whitespace-delimited file and post it here, I will compute the parameter variances.
Required fields in each record:
r1 r2 r3 b1 b2 b3 rsu bsu
... where rsu & bsu are the unpenalized alliance scores