<p>一般线性模型是不是就指线性回归？或者说是广义线性模型的特殊形式?</p><p>不知道用什么命令做这个。是用regress还是GLM？</p><p>在SPSS里面general linear model, generalized linear model,regress这几个命令都有，有些不明白</p><p></p><p>这是statalist里面的一段话，请教高手</p><p></p><p>I can't say what you need. <br/><br/>Historically the abbreviation -glim- (or more<br/>precisely GLIM) was used for software implementing <br/>generalized l.m.s (sensu Nelder and Wedderburn). <br/><br/>My guess is that -glm- was used in Stata as a <br/>slightly different name precisely to avoid any <br/>inference (implication, too) that -glm- matched <br/>GLIM one-to-one, which it certainly did not. <br/><br/>But if Joe Hilbe is watching he can beat guessing<br/>and tell us why he chose that name. <br/><br/>Nick <br/>n.j.cox@durham.ac.uk <br/><br/>Ricardo Ovaldia<br/><br/>  <br/>> Thank you Nick. I new that generalized linear models<br/>> were not what I need it. Using the acronym "GLM" for<br/>> both general and generalized is very confusing. I<br/>> think that maybe -mvreg- is what I really need.<br/><br/>  <br/>Nick Cox<br/><br/>  <br/>> > Usually, -regress- or -glm-, except that <br/>> > neither captures absolutely all the <br/>> > functionality that some authors associate <br/>> > with GLM. Note that general linear models<br/>> > and generalised linear models are not <br/>> > the same families. <br/><br/>  <br/>Ricardo Ovaldia<br/><br/>> > > What command can I use to fit a General linear<br/>> > Model in Stata?<br/></p>

[此贴子已经被作者于2008-8-28 11:07:33编辑过]