The first part of the paper is concerned with a brief description of the lambda-continuum of inductive methods due to R. Carnap. In the second part an approximate representation of real numbers through binary successions allows an interpretation of the same like Q-predicates of a first order monadic language LpN. The compatibility between the lambda-continuum and the probabilistic structure of a random sampling without replacement from a finite population is proved and through the C-mean estimator definition the class F-lambda of the inductive estimate functions is derived. The variable Y_s+1 depicts the value (Y) of the individual not contained in the sample e_Q. The class F-lambda is examined with reference to Y_s+I estimation through the mean square error: the theoretical results are useful in estimator selection. In particular manner the “straight rule": (lambda=0) is applicable in narrow circumstances.
Il lambda-continuo dei metodi induttivi di R. Carnap: un'applicazione nell'ambito della teoria della stima nel campionamento in blocco da popolazioni finite
GUSEO, RENATO
1980
Abstract
The first part of the paper is concerned with a brief description of the lambda-continuum of inductive methods due to R. Carnap. In the second part an approximate representation of real numbers through binary successions allows an interpretation of the same like Q-predicates of a first order monadic language LpN. The compatibility between the lambda-continuum and the probabilistic structure of a random sampling without replacement from a finite population is proved and through the C-mean estimator definition the class F-lambda of the inductive estimate functions is derived. The variable Y_s+1 depicts the value (Y) of the individual not contained in the sample e_Q. The class F-lambda is examined with reference to Y_s+I estimation through the mean square error: the theoretical results are useful in estimator selection. In particular manner the “straight rule": (lambda=0) is applicable in narrow circumstances.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.