Hey Bro(ettes),
I need to create a synthetic data set with realistic ages for seniors. I know the mean and standard deviation of the population. Obviously, I don't want to assume that the ages are normally distributed for people aged 65 and older.
Does anyone know what distribution is commonly used for generating realistic age data?
Thanks.
WTF? The actual data is public and freely available online
I only know the mean and standard deviation. The actual data is in a secure data facility that I don't have access to. I'm also trying to make things simple.
Fair enough, I was hoping to tell my RA to use one line of code instead of 50.
WTF? The actual data is public and freely available onlineI only know the mean and standard deviation. The actual data is in a secure data facility that I don't have access to. I'm also trying to make things simple.
Fair enough, I was hoping to tell my RA to use one line of code instead of 50.
Use an Erlang distribution and the mean and standard deviation you already have.
WTF? The actual data is public and freely available onlineI only know the mean and standard deviation. The actual data is in a secure data facility that I don't have access to. I'm also trying to make things simple.
Fair enough, I was hoping to tell my RA to use one line of code instead of 50.
There is no possible way you have even a high school education.
Oh my god…
There is no possible way you have even a high school education.
Oh my god…Tenured professor bro, come at me.
You don’t seem to grasp that downloading the true distribution would give you the true distribution without having to write a single line of code.
Honestly, if you’re a tenured prof, then you are possibly suffering from dementia.
You don’t seem to grasp that downloading the true distribution would give you the true distribution without having to write a single line of code.
Honestly, if you’re a tenured prof, then you are possibly suffering from dementia.
I need to generate micro-data where age is just one of many covariates. So the true distribution would allow me to generate this variable, but with lots of replace commands.
You don’t seem to grasp that downloading the true distribution would give you the true distribution without having to write a single line of code.
Honestly, if you’re a tenured prof, then you are possibly suffering from dementia.I need to generate micro-data where age is just one of many covariates. So the true distribution would allow me to generate this variable, but with lots of replace commands.
You're really not that bright if you cant figure out how to do that with a loop command.
WTF? The actual data is public and freely available onlineI only know the mean and standard deviation. The actual data is in a secure data facility that I don't have access to. I'm also trying to make things simple.
Fair enough, I was hoping to tell my RA to use one line of code instead of 50.
Use an Erlang distribution and the mean and standard deviation you already have.
This is the only helpful answer -- I was going to suggest gamma or any other two-parameter distribution supported on the positive real line. Solve for the parameters given the observed mean and variance.