Ravi ★ India, 20080602 12:39 (5895 d 03:39 ago) Posting: # 1897 Views: 7,801 

Dear all, My question is what is the basic difference between Type 1, 2 and 3 sum of square. Please reply as soon as possible. Ravi — Thanks & Regards Ravi Pandey 
Ohlbe ★★★ France, 20080602 13:57 (5895 d 02:21 ago) @ Ravi Posting: # 1898 Views: 6,158 

Dear Ravi, ❝ Please reply as soon as possible. If you are in such a hurry, why didn't you use the search function of the forum in the first place ? You would have found this thread, for instance. Regards Ohlbe 
martin ★★ Austria, 20080604 00:40 (5893 d 15:38 ago) @ Ravi Posting: # 1904 Views: 6,216 

dear ravi ! there is often a misunderstanding regarding sum of squares. there is in fact only one sum of squares, the improvement of the sum of squares for the alternative hypothesis over the sum of squares for the null hypothesis. With a tool like R (www.Rproject.org) you simple fit model1 then fit model2 and compare fits using anova(model1, model2) .Refer to http://www.matthewckeller.com/TypeIorTypeIIISS.pdf for more details. best regards martin 
Helmut ★★★ Vienna, Austria, 20080605 02:25 (5892 d 13:53 ago) @ martin Posting: # 1920 Views: 6,288 

Dear Martin, thanx for the link  which lead me to a quite interesting paper by Bill Venables from 2000. I like the quote on page 12: [...] SAS seems to be to statistical computing what Microsoft is to personal computing [...] The Type III stuff starts at page 13... — Diftor heh smusma 🖖🏼 Довге життя Україна! _{} Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes 