Analysis of Data for Measuring Food Availability, Access and Nutritional Status (Topic 18)

Training Material
Subtitle: 
[Multiple Causes of Food Insecurity: Multiple Regression]

How can one estimate and assess the differential importance of the varied causes of food insecurity? Focus on one single explanatory variable as done in simple linear regression analysis would be misleading. In real life, other things do not really remain the same though economic theory postulates relationships subject to ‘ceteris paribus’. How do we accommodate such factors in regression analysis? Does multiple regression take care of ‘ceteris paribus’? How should one specify and estimate a multiple regression model? What are its the advantages and disadvantages in an empirical context? What are the pitfalls in multiple regression and what are the checks to do? What is the relationship between partial regression coefficients and zero-order correlation coefficients? What is meant by first-order or partial correlation coefficients? How does one measure the association between variables after netting out the impact of excluded variables? How does one verify if inclusion of additional explanatory variables would improve the explanatory power of the regression model? How does one adjust for degrees of freedom while assessing the explanatory power of the model? What are the practical applications of such measures and their properties in empirical policy analysis? These issues would be discussed under this topic.

Download documentSize
Analysis of Data for Measuring Food Availability, Access and Nutritio...128.12 KB
Analysis of Data for Measuring Food Availability, Access and Nutritio...335.67 KB
Date and language
Jan 2012
English
Personal Author
Location
Physical Location: 
FPMU Documentation Center
Other information
Form: 
Slides
Form: 
Web resource

Share this