作者: A. Wünsche , W. Näther
DOI: 10.1016/S0165-0114(02)00103-3
关键词:
摘要: Let X˜,Y˜ be two convex fuzzy random variables on Rn. Using a suitable metric we prove that the conditional expectation E(Y˜ | X˜) is best approximation of Y˜ by measurable functions X˜. This generalizes analogous and well known property for real variables. A further topic linear function In special cases use Hukuhara's difference between sets, obtain formulas which are to classical structure. Contrary fact, however, Gaussian in general does not coincide with regression function.