| home | library | feedback |
Dependence modeling means finding the model of the probabilistic dependences of the variables. The Bayesian approach gives us a way to recognize a good model when we find one.
Dependence modeling means finding the model of the probabilistic dependences of the variables. It tries to find dependencies between all the variables in your data. Dependencies can be used to speculate about causalities that might cause them. Besides revealing the structure of the domain of your data, dependency models can be used to infer probabilities of any set of variables given any (other) set of variables. This will lead to the game where you can interactively study your model by probing it.
Being Bayesian gives us a way to recognize a good model when we see one. Simply stated, a good dependence model is one with a high probability. However, it takes a Bayesian approach to speak about the probability of the dependencies. "Classical" (frequentist) statistician is not allowed to speak about probabilities of dependencies. Why? It is a philosophical issue, and related to the question of how to use the probability theory to answer our question. It is also an issue that can be debated. If interested, you can find more information from the texts in the B-Course library.
» More about dependency models
» More about Bayesians and dependency models
» About statistical dependencies
|
|
|
|
| B-Course, version 2.0.0 | |