R-square value of Probability plot
古いコメントを表示
Is there an easy way to find the R-squared value for a probability plot which I plotted using probplot() in MATLAB?
3 件のコメント
dpb
2013 年 8 月 13 日
Doesn't appear that it returns anything but the handle(s) to the lines drawn; not the fitted parameters.
Looks like you'll need to estimate the coefficients and the compute the SSE and SST
doc normfit % for details
the cyclist
2013 年 8 月 13 日
dpb, any reason why you put this as a comment, rather than an answer that could be accepted?
dpb
2013 年 8 月 13 日
Wasn't sure it qualified as enough detail...but per your suggestion, done. :)
回答 (1 件)
dpb
2013 年 8 月 13 日
0 投票
Doesn't appear that probplot returns anything but the handle(s) to the lines drawn; not the fitted parameters.
You'll needs must estimate the coefficients and the compute the SSE and SST resulting therefrom.
doc normfit % for details
2 件のコメント
Lisha
2013 年 8 月 19 日
How did you do the plot, then?
If you don't have the data then you have no way to compute the residuals and hence no way to compute SSE (or SST, even).
ADDENDUM:
Perhaps you mean the ancillary statistics beyond the estimated parameters aren't available instead of the underlying data? In that case, my misinterpretation and you then use the definitions of SSE and SST to compute from the residuals which you can obtain by using normpdf with the underlying data and the estimated/fitted mean, std dev. That help?
カテゴリ
ヘルプ センター および File Exchange で Exploration and Visualization についてさらに検索
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!