My reflexion originates from the notion of covering model with nice estimation properties. This happens to be particularly useful in studies for which the outcome of interest is measured only on a selected population. Let X be an explanatory variable of the outcome Y and Z be a variable that drives the selection mechanism. Two situations can occur, one in which Y is defined for all units in the population but observed only for Z taking on some particular values and the other in which Y is defined only for those units with Z taking on particular values. I shall discuss the first one and build an example to summarize what I mean, and to clarify my comment.
Comments on: Sequence of regressions and their independences
STANGHELLINI, Elena
2012
Abstract
My reflexion originates from the notion of covering model with nice estimation properties. This happens to be particularly useful in studies for which the outcome of interest is measured only on a selected population. Let X be an explanatory variable of the outcome Y and Z be a variable that drives the selection mechanism. Two situations can occur, one in which Y is defined for all units in the population but observed only for Z taking on some particular values and the other in which Y is defined only for those units with Z taking on particular values. I shall discuss the first one and build an example to summarize what I mean, and to clarify my comment.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
