tag:blogger.com,1999:blog-37324607.post3860092978868208126..comments2023-11-03T08:31:23.698-04:00Comments on Data Mining in MATLAB: Linear Regression in MATLABWill Dwinnellhttp://www.blogger.com/profile/03379859054257561952noreply@blogger.comBlogger7125tag:blogger.com,1999:blog-37324607.post-75190123959404033682009-09-03T22:00:15.781-04:002009-09-03T22:00:15.781-04:00Hi,
I am doing multiple regression using matlab ...Hi, <br /><br />I am doing multiple regression using matlab with three dependent variables. Matlab is spitting out only 1 p-value or strictly speaking, the F-statistic. How can i get the p-values corresponding to all of the dependent variables? <br /><br />Thanks.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-84186546511782390502008-12-03T07:02:00.000-05:002008-12-03T07:02:00.000-05:00dear Will Dwinnell,many compliments for you work o...dear Will Dwinnell,<BR/>many compliments for you work on MATLAB. <BR/><BR/>I'd like to know if there is a solution/workaround form my statistical problem.<BR/><BR/>I'd like to understand, just as overview, how I can solve this problem: I know yi(outcome) and wi(initial weights) and xi (values) but I don't know the f(x,w)<BR/><BR/>My function to predict is:<BR/><BR/>y = f(w1*x1,w2*x2,...,w39*x39)<BR/><BR/>I'd like to minimize error for predicting yi and find the final weight.<BR/><BR/>I'm wondering if is a Optimization problem: http://www.mathworks.com/matlabcentral/fileexchange/8553<BR/><BR/>I'm a statistician and I'd think to solve this example with nonlinear regression with Bound constraint, but I ask you if with Matlab can solve it better.<BR/><BR/>Any kind of suggestions will be really appreciated.<BR/><BR/>Thanks in advanceCristianohttps://www.blogger.com/profile/12013779853484062870noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-17711938079975665532008-09-21T05:44:00.000-04:002008-09-21T05:44:00.000-04:00[B,BINT,R,RINT,STATS] = REGRESS(Y,X) returns a vec...[B,BINT,R,RINT,STATS] = REGRESS(Y,X) returns a vector STATS containing, in<BR/> the following order, the R-square statistic, the F statistic and p value<BR/> for the full model, and an estimate of the error variance.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-28581812494875820662008-01-04T19:35:00.000-05:002008-01-04T19:35:00.000-05:00I never use those statistics, so I do not have any...I never use those statistics, so I do not have any code immediately handy. The <I>regress</I> function in the Statistics Toolbox will generate a number of these diagnostics, and it should not be hard to create one's own calculations using MATLAB's built-in functions, like <I>var</I>.Will Dwinnellhttps://www.blogger.com/profile/03379859054257561952noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-58070206024896321012008-01-04T13:28:00.000-05:002008-01-04T13:28:00.000-05:00How does one attain simple diagnostic statistics a...How does one attain simple diagnostic statistics about the multiple regression, such as: <BR/>-standard error<BR/>-t statistic<BR/>-P-value<BR/>-confidence interval<BR/>-r square<BR/>-adjusted r square<BR/><BR/>in matlab? <BR/>These are available in Excel with hte click of a button, but I'm positive Matlab should be way better than Excel :PAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-37324607.post-27364872703474497092007-04-25T20:41:00.000-04:002007-04-25T20:41:00.000-04:00I was not aware of that particular title. It look...I was not aware of that particular title. It looks like it's time to brush up on my linear algebra!Will Dwinnellhttps://www.blogger.com/profile/03379859054257561952noreply@blogger.comtag:blogger.com,1999:blog-37324607.post-82972572571189498232007-04-25T02:43:00.000-04:002007-04-25T02:43:00.000-04:00Hello Will,This comment is not related to your pos...Hello Will,<BR/><BR/>This comment is not related to your post but it may be of interest to you. I've just found a new book on amazon about <A HREF="http://www.amazon.com/gp/product/0898716268/ref=pe_pe_5050_5507230_pe_snp_268" REL="nofollow">data mining and matlab</A>. Perhaps you already know it.Sandro Saittahttps://www.blogger.com/profile/17682082649770875583noreply@blogger.com