Showing 1 - 5 of 5
A common problem in applied regression analysis is that covariate values may be missing for some observations but imputed values may be available. This situation generates a trade-off between bias and precision: the complete cases are often disarmingly few, but replacing the missing observations...
Persistent link: https://www.econbiz.de/10010640491
A common problem in applied regression analysis is that covariate values may be missing for some observations but imputed values may be available. This situation generates a trade-off between bias and precision: the complete cases are often disarmingly few, but replacing the missing observations...
Persistent link: https://www.econbiz.de/10010821074
represented by the decrease in entropy of his beliefs, regardless of his preferences, initial wealth, or investment problem. We …
Persistent link: https://www.econbiz.de/10010930175
We present an axiomatization of the entropy of capacities defined on set systems which are not necessarily the whole … power set, but satisfy a condition of regularity. This entropy encompasses the definition of Marichal and Roubens for the … entropy of capacities. Its axiomatization is in the spirit of the one of Faddeev for the classical Shannon entropy. In …
Persistent link: https://www.econbiz.de/10010750597
entropy, namely the degree of disorder within a system. In this paper we assume that "complete" and "accurate" real …-time information sharing concerning product and inventory levels through RFID reduces entropy and limits its effects on the supply … informational entropy in the supply chain. …
Persistent link: https://www.econbiz.de/10011025782