The "statistical experiment"-equivalence for prior distributions
نویسندگان
چکیده
منابع مشابه
A Statistical Test for Joint Distributions Equivalence
We provide a distribution-free test that can be used to determine whether any two joint distributions p and q are statistically different by inspection of a large enough set of samples. Following recent efforts from Long et al. [1], we rely on joint kernel distribution embedding to extend the kernel two-sample test of Gretton et al. [2] to the case of joint probability distributions. Our main r...
متن کاملPrior Distributions for ELISA Analyses
Based on the design of the ELISA calibration in the comparability study CCQM-P58, three measured signals are simulated at each of the 11 known concentrations (100, 50, . . . , 0.195, 0) ng/ml. These calibration data are simulated according to Model (1.1) using the parameters β = (1.466, 0.1024, 67.93, 1.361), a = 3.933 × 10, c = 9.710 × 10, and are listed in Table 1. Based on the comparability ...
متن کاملCompatible Prior Distributions
We investigate two approaches to constructing compatible prior laws over alternative models: ‘projection’ and ‘conditioning’. Each of these is shown to require additional inputs. We suggest that these can be chosen in a natural way in each case, leading to ‘Kullback-Leibler projection’ and ‘Jeffreys conditioning’. We recommend the former for the case of coexisting models, and the latter for com...
متن کاملEnsemble equivalence for non-Boltzmannian distributions
We discuss the possibility of using generalized canonical distributions, i.e., using other factors than expð bEÞ, in order to compute the equilibrium properties of physical systems. It will be shown that some other choices can, in certain cases, lead to a simpler calculation of those properties. The corresponding equivalence between the canonical and the generalized canonical distributions is d...
متن کاملFlexible Prior Distributions for Deep Generative Models
We consider the problem of training generative models with deep neural networks as generators, i.e. to map latent codes to data points. Whereas the dominant paradigm combines simple priors over codes with complex deterministic models, we argue that it might be advantageous to use more flexible code distributions. We demonstrate how these distributions can be induced directly from the data. The ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Bulletin of the Belgian Mathematical Society - Simon Stevin
سال: 1998
ISSN: 1370-1444
DOI: 10.36045/bbms/1103211557