>
Fa   |   Ar   |   En
   ExtremeBounds: Extreme bounds analysis in R  
   
نویسنده hlavac m.
منبع journal of statistical software - 2016 - دوره : 72 - شماره : 0
چکیده    This article introduces the r package extremebounds to perform extreme bounds analysis (eba),a sensitivity test that examines how robustly the dependent variable of a regression model is related to a variety of possible determinants. extremebounds supports leamer’s eba that focuses on the upper and lower extreme bounds of regression coefficients,as well as sala-i-martin’s eba which considers their entire distribution. in contrast to existing alternatives,it can estimate models of a variety of user-defined sizes,use regression models other than ordinary least squares,incorporate non-linearities in the model specification,and apply custom weights and standard errors. to alleviate concerns about the multicollinearity and conceptual overlap of examined variables,extremebounds allows users to specify sets of mutually exclusive variables,and can restrict the analysis to coefficients from regression models that yield a variance inflation factor within a prespecified limit. © 2016,american statistical association. all rights reserved.
کلیدواژه Extreme bounds analysis; R; Regression; Robustness; Sensitivity
آدرس harvard university,john f. kennedy school of government,79 john f. kennedy street,cambridge,ma 02138, United States
 
     
   
Authors
  
 
 

Copyright 2023
Islamic World Science Citation Center
All Rights Reserved