Qian, G. and Kunsch, H. (1996). Some Notes On Rissanen's Stochastic Complexity, preprint. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Master Thesis - Lanterman (Correct)
....of data under the model. If the square root of the Fisher information is not integrable, then we can integrate over subsets of the full space, and use Rissanen s log to denote the subset. Several examples of this are given in Section V of [33] this is presented in a general form in Eq. 7 of [102]. We choose an increasing sequences X (1) ae X (2) ae Delta Delta Delta of bounded open subsets converging to X , and take the stochastic complexity to be Gamma ln p(yjm) GammaL(yjx ML ) Gamma d 2 ln 2 ln Z X (a) q det I L ( xjm)d x ln (a) c nat ; A.19) where a is the ....
....to different expressions for the stochastic complexity. In [33] Rissanen chooses them to ease the computation of the integral in (A.19) Addressing this issue rigorously should be a subject of future research. Remark: Using different assumptions and approximations than Rissanen, Qian and Kunch [102] derive a different approximation for stochastic complexity. We mention it for completeness and will not consider it further here. A.3 Examples of Stochastic Complexity and MDL This chapter concludes with some examples to help illustrate the concepts explained above. The first comes from the ....
G. Qian and H.R. Kunsch, "Some notes on Rissanen's stochastic complexity," IEEE Trans. on Information Theory, vol. 44, no. 2, pp. 782--896, March 1988.
Schwarz, Wallace, and Rissanen: Intertwining Themes in Theories.. - Lanterman (1999) (2 citations) (Correct)
....complexity of data under the model. If the square root of the Fisher information is not integrable, then we can integrate over subsets of the full space, and use log to denote the subset. Several examples of this are given by Rissanen (1996, Section V) this is presented in a general form by Qian Kunsch (1998, Eq. 7) We choose an increasing sequence X (1) ae X (2) ae Delta Delta Delta of bounded open subsets converging to X , and take the stochastic complexity to be Gamma ln p(yjm) GammaL(yjx ML ) Gamma d 2 ln 2 ln Z X (a) p det I L ( xjm)d x ln (a) ln(c) 71) where a is ....
....sequence will lead to different expressions for the stochastic complexity. Rissanen (1996) chooses them to ease the computation of the integral in (71) Addressing this issue rigorously should be a subject of future research. Remark: Using different assumptions and approximations than Rissanen, Qian Kunch (1998) derive a different approximation for stochastic complexity. We mention it for completeness and will not consider it further here. 9 Conclusions This paper has detailed a wide variety of model selection methods, and attempted to alleviate some common misconceptions about them. Some of the ....
Qian, G. & Kunsch, H. (1998). Some notes on Rissanen's stochastic complexity. IEEE Trans. on Information Theory, 44(2):782--896.
On Model Selection in Robust Linear Regression - Qian, Künsch (1996) (3 citations) Self-citation (Qian Kunsch) (Correct)
No context found.
Qian, G. and Kunsch, H. (1996). Some Notes On Rissanen's Stochastic Complexity, preprint.
Computing Minimum Description Length for Robust Linear Regression.. - Qian (Correct)
No context found.
Qian, G., and K#unsch, H. #1998a#. Some notes on Rissanen's stochastic complexity. IEEE Trans. Inform. Theory. 44, 782-786.
Minimum Description Length Understanding of Infrared Scenes - Lanterman (Correct)
No context found.
G. Qian and H. Kunsch, "Some notes on Rissanen's stochastic complexity," IEEE Trans. on Information Theory 44, pp. 782--896, March 1988.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC