Daffodil International University

Faculties and Departments => Teaching & Research Forum => Topic started by: anam on November 12, 2016, 10:36:59 PM

Title: Regression Analysis
Post by: anam on November 12, 2016, 10:36:59 PM

Is it possible to have a (multiple) regression equation with two or more dependent variables?

Title: Re: Regression Analysis
Post by: saratasneem on November 14, 2016, 02:02:09 PM

I have the same question. As I have searched the literature (a little bit), it has been observed that , defferent equations are designed and estimated for different dependent variables but same independent variables.

Title: Re: Regression Analysis
Post by: imranmahmud on November 15, 2016, 11:51:08 PM

Of course it is possible. Suppose u have A,B->C,D. A and B independent, C and D dependent

Multiple regression is an extension of simple linear regression in which more than one independent variable (X) is used to predict a single dependent variable (Y) (Original Definition). By noting that you will get two equations for two dependent variables.

 in SPSS, you can do multiple regression from A,B->C and A,B->D.

This is the calculation technique. But if you want to test or run this model together than you must use PLS-SEM or CB-SEM. For PLS-SEM you can use smartPLS software, for CB-SEM , you can use AMOS.

Or you can do 2nd order , by merging C and D together. Then C and D will become E and you can test A,B->E
Usually , we do not put any equation because of the standard error. We just test the positive or negative relationship, its significance, Variance (R square) and effect size (f sqaure) .

Check the attached paper for multiple dependent variables and multiple regression.


Thanks

Title: Re: Regression Analysis
Post by: Saujanna Jafreen on November 26, 2016, 03:01:57 PM

yes it is.... :)

Title: Re: Regression Analysis
Post by: Bipasha Matin on November 27, 2016, 04:21:39 PM

helpful discussion in comment section

Title: Re: Regression Analysis
Post by: Nujhat Anjum on December 12, 2016, 01:52:02 PM

Thanks for sharing.

Title: Re: Regression Analysis
Post by: Nujhat Anjum on December 12, 2016, 01:53:30 PM

Thanks for sharing.

Title: Re: Regression Analysis
Post by: anam on April 09, 2017, 10:20:00 AM

Dear Imran Mahmud sir, please mention the methodology or any link for solving multiple regression where two independent variable connected with same dependent variable.

Thank you sir for your response.

Title: Re: Regression Analysis
Post by: Nusrat Nargis on April 10, 2017, 11:36:57 AM

good one.

Title: Re: Regression Analysis
Post by: imranmahmud on April 10, 2017, 02:27:12 PM

Dear Anam sir,

Please see my paper, two IV (H5 and H6) to 1 DV and also 5 IV  (H2, H4, H7, H8,H9)to 1 DV.
Thanks

Title: Re: Regression Analysis
Post by: imranmahmud on April 10, 2017, 02:44:53 PM

Dear Anam sir, in that paper though our DV is Actual Use, but we need to run 3 regression if you use spss for my model. In SEM-PLS, you can run all the three regression together.  The name of IV and DV is nowadays changed to exogenous (where arrows are going out) and endogenous (where arrows are coming in) variables. When no arrows are going out from any variable, then that is our DV.

Title: Re: Regression Analysis
Post by: Arfuna Khatun on April 17, 2017, 01:04:15 PM

yes

Title: Re: Regression Analysis
Post by: hassan on April 20, 2017, 07:59:38 AM

Nice sharing

Title: Re: Regression Analysis
Post by: Md.Shahjalal Talukder on April 20, 2017, 08:46:10 AM

Thanks

Title: Re: Regression Analysis
Post by: anam on April 21, 2017, 10:33:57 AM

Dear Imran sir please share the link of your paper if possible.

Title: Re: Regression Analysis
Post by: imranmahmud on April 21, 2017, 11:08:06 AM

Dear Sir,

Here is the link. http://www.sciencedirect.com/science/article/pii/S0736585316301642

Paper is attached in my previous post.

Title: Re: Regression Analysis
Post by: imranmahmud on April 21, 2017, 11:10:58 AM

Anam Sir, I emailed you some papers.

Title: Re: Regression Analysis
Post by: anam on April 22, 2017, 11:18:09 PM

Thank you sir. I got your paper.

Title: Re: Regression Analysis
Post by: Nargis Akter on April 23, 2017, 05:00:50 PM

Thanks