Logit Model xtmelogit (World Value Survey data)

[Anna Petrova]

Hi everyone,

I am writing my thesis with the topic "Determinants of charitable giving into generational differences" and need your help with coding. Will be very appreciate for our help!

My model is based on the Wave 6 of World Value Survey (pooled data).

I would like to have 2 cluster of people - Millennials (born year: 1980 - 1995) and Baby Boomers (born year: 1946 -1963) and run the model using xtmelogit to see the significance of the different individual characteristics across each generational group.

Can you advice please which model can better describe this question (I was advised to use xtmelogit) but maybe there is a better solution and help with creating of these 2 generational clusters in STATA.

I have generated dependent variable -binary variable and independent variables (in .do file). I have also created the binary variable for age but I think it is wrong if I want to run the model for each cluster separately.... I am not sure if I need to include some variables to control for country effect. (as a base I have used the paper attached - also based on WVS and use ordered login model/ logit model by clusters)

Will be very appreciate for all your advices!

Thank you advance!

files are attached

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Also, many folks will not open files due to concerns about viruses.

If you only have one kind of panel, then xtlogit will do the job. If you want to include individual characteristics (that don't vary over time within individuals), then you can't do a conventional fixed effect estimator. You may want to look at xthybrid and the related Stata Journal articles.

What I would normally expect this analysis to be is to set the panel as individuals, and then use a factor variable (interaction) to estimate parameters for both Millennials and Baby Boomers in the same logit. This makes testing parameter differences easy and let's you use the margins command after estimation.