Variable length & Stepwise regression

Question

Yaojiayin 2014-9-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/155300-variable-length-stepwise-regression

评论： Yaojiayin 2014-9-19

采纳的回答： Roger Wohlwend

Araa06.txt

在 MATLAB Online 中打开

I want to create a dataset

A = dataset('File',Araa06.txt');

Matlab returens errors:

Misuse dataset/readFile (line 207) Variable lengths must all be the same.
Error dataset (line 351) a = readFile(a,fileArg,otherArgs)
What could be the problem?

If I want to use dataset A for stepwise regression

 mdl=stepwiselm(A,'Predictor variables','DEM','grad','Response variable','Yield')

What is the different between this one and

stepwise

If I just want to do stepwise regression, do I really need to build a model？

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Roger Wohlwend 2014-9-19

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/155300-variable-length-stepwise-regression#answer_152219

If you copy the contents of your txt-file and paste it into an Excel file, you see that not all numbers are separated by a tab. Some are separated by blanks. I guess that is why Matlab cannot read the txt file properly. It expects all numbers to be separated by tabs. So you have to fix the txt file first.

For stepwise regression you can use either stepwiselm or stepwise. They are basically the same. The latter offers you a GUI, the first is a class. I would recommend stepwiselm. You don't need a GUI. Just work with the class.

1 个评论
显示 -1更早的评论隐藏 -1更早的评论

Yaojiayin 2014-9-19

Thanks for your reply! I fixed the problem by just copy the first row(name of variables) in the original data file (in form of .txt) to Araa06, and it worked (for some reason). Then I did stepwiselm and wrote a report to my teacher.

Thanks again for answering my question!

请先登录，再进行评论。