Statistics Help @ Talk Stats Forum - Regression Analysis
http://www.talkstats.com/
Linear regression, linear models, nonlinear regressionenThu, 23 Mar 2017 09:41:18 GMTvBulletin60http://www.talkstats.com/images/misc/rss.pngStatistics Help @ Talk Stats Forum - Regression Analysis
http://www.talkstats.com/
I need help with my regression analysis for research capstone.
http://www.talkstats.com/showthread.php/69042-I-need-help-with-my-regression-analysis-for-research-capstone.?goto=newpost
Wed, 22 Mar 2017 19:59:48 GMTHi all,

I'm new to regressions and also new to posting here.

My topic is: How have E-books Affected College Textbook Prices?

My hypothesis is:
H0: The CPI for college textbooks will fall with the introduction of the e-book because increased innovation and competition causes an increase in supply causing prices to fall.

H1: The CPI for college textbooks is unaffected by the introduction of the e-book.
(I can change my alternative and probably will since my r2 is >80% as of now...)

I have basically no idea how to do regressions so I used a online helper and I came up with this:

My dependent variable (x) is the CPI for College Textbooks by the BLS from 2001 to 2016 separated by year.

My independent variable (y) is the # of ebooks sold from 2001 to 2016 separated by year.

What I'm trying to figure out is what does all of this mean? My p val is <.0001 which is good I think but I need help putting what all of this info means in words?

Also, I have other variable (control variables I'm guessing?) such as the Consumer Price Index for College Tuition and Fees, GDP, and Higher ED Enrollment among other things and I want them to be a part of the regression. The prob is I don't know how to add them into it!!!

Any help is appreciated.

Thank you so much an advance!

PS I have the data in excel and can include also.

]]>Regression Analysistlowehttp://www.talkstats.com/showthread.php/69042-I-need-help-with-my-regression-analysis-for-research-capstone.<![CDATA[What do you do it you can't increase X by one level while holding all other X constan]]>
http://www.talkstats.com/showthread.php/69041-What-do-you-do-it-you-can-t-increase-X-by-one-level-while-holding-all-other-X-constan?goto=newpost
Wed, 22 Mar 2017 17:36:10 GMTClassical regression defines the slope as the change in Y for a one unit change in X holding all other X constant (or so I believe). :p What happens...Classical regression defines the slope as the change in Y for a one unit change in X holding all other X constant (or so I believe). :p What happens if its impossible to hold all other X constant when you change a given X. In some cases, substantively, if you change a specific X another X has to change, one is inherently linked to the other.

Does the slope definition change then, or is there a technique that addresses this?
]]>Regression Analysisnoetsihttp://www.talkstats.com/showthread.php/69041-What-do-you-do-it-you-can-t-increase-X-by-one-level-while-holding-all-other-X-constanMultiple regression analysis: linear or logistic?
http://www.talkstats.com/showthread.php/69035-Multiple-regression-analysis-linear-or-logistic?goto=newpost
Tue, 21 Mar 2017 18:51:28 GMTHello!
I have a question regarding multiple regression analysis: for numerical variable I should use linear regression and for dichotomous...Hello!

I have a question regarding multiple regression analysis: for numerical variable I should use linear regression and for dichotomous variables I should use logistic regression. But if I have numerical and dichotomous variables, which regression should I use? Thank you!
]]>Regression Analysislacrihttp://www.talkstats.com/showthread.php/69035-Multiple-regression-analysis-linear-or-logistic