48 visualizaciones (últimos 30 días)
Tom Lane el 8 de Mayo de 2012
You can get R^2 (multiple correlation) from various Statistics Toolbox routines. If you don't have the toolbox or you need to do it yourself, you could use backslash to carry out a regression, then compare the sum of squared deviations from the mean of both the y variable and the residuals from the regression. Alternatively you could use corrcoef to compute the correlation between the y variable and the fitted values.
My guess is that a book on regression will give you a better explanation than that wikipedia page.