Hello,
I estimated a beta regression with dependent variable that falls strictly within the range of 0 and 1. Are there any assumptions that I must test after estimation? Does assumptions of normality, absence of autocorrelation, and heteroskedasticity matter in this case? Is it sufficient to just check for specification errors after estimation like in fractional regression model?

Thank you