Statistics Help @ Talk Stats Forum - Regression Analysis
http://www.talkstats.com/
Linear regression, linear models, nonlinear regressionenFri, 24 Nov 2017 23:05:13 GMTvBulletin60http://www.talkstats.com/images/misc/rss.pngStatistics Help @ Talk Stats Forum - Regression Analysis
http://www.talkstats.com/
Multilevel logistic regression with rare events
http://www.talkstats.com/showthread.php/70832-Multilevel-logistic-regression-with-rare-events?goto=newpost
Fri, 24 Nov 2017 17:14:48 GMTHi,

I've conducted a survey with a binary dependent variable that I assessed twice. My sample is small (n=160) and I got one answer rarely (0=6; 1=154).

I'm wondering which statistical test to use. I know that there is multilevel logistic models to account for the nested data structure and that there is, for example, penalized maximum likelihood for the small sample/rare events. Is there a combination of both?

I'm using Stata14.

Best,
akischo
]]>Regression Analysisakischohttp://www.talkstats.com/showthread.php/70832-Multilevel-logistic-regression-with-rare-eventsMultiple Regression
http://www.talkstats.com/showthread.php/70827-Multiple-Regression?goto=newpost
Fri, 24 Nov 2017 04:15:17 GMTI am trying to understand the relationship between livestockunits(X) and gross margin earnings(Y). I have generated different models using multi...I am trying to understand the relationship between livestockunits(X) and gross margin earnings(Y). I have generated different models using multi variables inaddition to my main independent variable.
In one of the multi Regression model, individual t-test of the main independent variable is not significant at 5%. But overall F-test of all variables is significant at 5%. Is this due to multicollinearity issue and Omitted Variable Bias? And R squared only explains 2% of the variation of Y. With my limited understanding, i have planned to add omitted variables and try to improve my variation? Is this correct way of doing it?
]]>Regression AnalysisTitushahhttp://www.talkstats.com/showthread.php/70827-Multiple-RegressionEnglish name for this index - Least Squares Method
http://www.talkstats.com/showthread.php/70806-English-name-for-this-index-Least-Squares-Method?goto=newpost
Wed, 22 Nov 2017 13:53:58 GMTHi all,

I was wondering what is the English name for the index I_2, used to evaluate the goodness of the fit:

I_2= SQRT[ SUM(Y_i-y_i)^2/N]/SUM(Y_i)/N,

where

Y_i is the ith value of the variable to be predicted and y_i=m*X_i+q is the predicted value of Y_i. N is the number of points (Y_i, X_i).

Many Thanks,
hannibal
]]>Regression Analysishannibalhttp://www.talkstats.com/showthread.php/70806-English-name-for-this-index-Least-Squares-Method