Aggregating Data by Subgroups in STATA

Using the collapse command to create aggregate data from individual-level data using frequency weights.
Visit my website for more videos:
davidbraudt.com...

[@mohamoudmohamedabdinoor5107]

Really you are the best teacher i have ever seen thank you very much, you tough me something than i don't understand deeply from my teacher.

[@Bakoji2]

I found the vedios very informative

[@1ivoleuppi]

i have an urgent question: im trying to make a longitudional file from multiple eurobarometer files. i want to measure the changes in issues mentioned as the most important problem over time. can anyone help me. i know so little about stata i dont even know what to look for

[@thomaskirkland3831]

David, how do you collapse dummy variables (1/0) or variables with numerical categories (1, 2, 3, 4)? I am assuming you can't take the mean of these variables.

[@Carlos___Rz]

You would probably just ignore the information given by dummy variables unless you actually want to aggregate by those dummy variables. eg by year and gender...

[@boscokasundu8494]

+Thomas Kirkland Hi you can actually create dummies from the a categorical variable ie Frequency of smoking with four options such as 1. Most often. 2. Rarely 3. Never by using tab freq, gen (most_often) command this will generate variables with the name most_often1, 2, 3 where 1=most_often, 2= rarely etc

[@DavidBraudt]

+Thomas Kirkland +Bosco Kasundu Thanks for replying to this Bosco! I apparently missed it a year back. Hopefully you found an answer back then Thomas.

[@anasiqueira5920]

David, how can I regress in loop by firm and save the coefficient as a variable in the dataset?

[@estheramoako1197]

how do you add the result to the original data set for further analysis?

[@DavidBraudt]

Use the merge command on the variable that was used to collapse the data. You will need to use either the m:1 or the 1:m version of merge. For example if you have the collapsed dataset in memory and are merging in the original data (let's assume a case is an individual) it would be a 1:m merge since there is one collapsed observation in the "master" dataset (Stata's language) that corresponds to many observations in the "using" dataset (i.e. the individual-level data you are merging back into the collapsed data). HTH.

[@desherilac9322]

Thank you very helpful :)

[@eyita_admas]

how do we collapse if our variables of interest dummy in nature

[@riddhimastakar]

Hi, did you figure out how to do this?

[@aung9211]

Thank you

[@krabbypatty6896]

thanksss!!!!