#censoreddata — Public Fediverse posts

Dealing with Censored Earnings in Register Data https://d.repec.org/n?u=RePEc:hal:journl:hal-05084494&r=&r=eur
"… In many register #data worldwide a relevant part of earnings and wealth is top-coded (right-censored), , can severely undermine its usefulness. For instance, how can #inequality and top earnings be credibly studied if the right tail of the earnings distribution is missing? Taking a distributional approach that is based on the semi-parametric modelling of the right tail being Pareto-like, we show how the missing tail can be successfully estimated using the administrative censored data.
… The presented distributional approach should not be confused with individuallevel imputations of #censoredData. The latter requires, in addition to the Pareto parameter, the assignment of a rank of to each censored observation. Due to the censoring, these ranks are unobserved… "
#Econometrics #PublicStatistics #economics

Dealing with Censored Earnings in Register Data https://d.repec.org/n?u=RePEc:hal:journl:hal-05084494&r=&r=eur
"… In many register #data worldwide a relevant part of earnings and wealth is top-coded (right-censored), , can severely undermine its usefulness. For instance, how can #inequality and top earnings be credibly studied if the right tail of the earnings distribution is missing? Taking a distributional approach that is based on the semi-parametric modelling of the right tail being Pareto-like, we show how the missing tail can be successfully estimated using the administrative censored data.
… The presented distributional approach should not be confused with individuallevel imputations of #censoredData. The latter requires, in addition to the Pareto parameter, the assignment of a rank of to each censored observation. Due to the censoring, these ranks are unobserved… "
#Econometrics #PublicStatistics #economics

Dealing with Censored Earnings in Register Data https://d.repec.org/n?u=RePEc:hal:journl:hal-05084494&r=&r=eur
"… In many register #data worldwide a relevant part of earnings and wealth is top-coded (right-censored), , can severely undermine its usefulness. For instance, how can #inequality and top earnings be credibly studied if the right tail of the earnings distribution is missing? Taking a distributional approach that is based on the semi-parametric modelling of the right tail being Pareto-like, we show how the missing tail can be successfully estimated using the administrative censored data.
… The presented distributional approach should not be confused with individuallevel imputations of #censoredData. The latter requires, in addition to the Pareto parameter, the assignment of a rank of to each censored observation. Due to the censoring, these ranks are unobserved… "
#Econometrics #PublicStatistics #economics

Dealing with Censored Earnings in Register Data https://d.repec.org/n?u=RePEc:hal:journl:hal-05084494&r=&r=eur
"… In many register #data worldwide a relevant part of earnings and wealth is top-coded (right-censored), , can severely undermine its usefulness. For instance, how can #inequality and top earnings be credibly studied if the right tail of the earnings distribution is missing? Taking a distributional approach that is based on the semi-parametric modelling of the right tail being Pareto-like, we show how the missing tail can be successfully estimated using the administrative censored data.
… The presented distributional approach should not be confused with individuallevel imputations of #censoredData. The latter requires, in addition to the Pareto parameter, the assignment of a rank of to each censored observation. Due to the censoring, these ranks are unobserved… "
#Econometrics #PublicStatistics #economics

Dealing with Censored Earnings in Register Data https://d.repec.org/n?u=RePEc:hal:journl:hal-05084494&r=&r=eur
"… In many register #data worldwide a relevant part of earnings and wealth is top-coded (right-censored), , can severely undermine its usefulness. For instance, how can #inequality and top earnings be credibly studied if the right tail of the earnings distribution is missing? Taking a distributional approach that is based on the semi-parametric modelling of the right tail being Pareto-like, we show how the missing tail can be successfully estimated using the administrative censored data.
… The presented distributional approach should not be confused with individuallevel imputations of #censoredData. The latter requires, in addition to the Pareto parameter, the assignment of a rank of to each censored observation. Due to the censoring, these ranks are unobserved… "
#Econometrics #PublicStatistics #economics