Financial markets: very noisy information processing Export

@article{citeulike:600939, abstract = {We report new results about the impact of noise on information processing with application to financial markets. These results quantify the trade-off between the amount of data and the noise level in the data. They also provide estimates for the performance of a learning system in terms of the noise level. We use these results to derive a method for detecting the change in market volatility from period to period. We successfully apply these results to the four major foreign exchange markets. The results hold for linear as well as nonlinear learning models and algorithms and for different noise models}, author = {Magdon-Ismail, M. and Nicholson, A. and Abu-Mostafa, Y. S.}, citeulike-article-id = {600939}, citeulike-linkout-0 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=726786}, comment = {Looks at effect of noise and training size on "stable" learning systems. Expected squared error grows linearly with variance, and 1/N with N=training size}, journal = {Proceedings of the IEEE}, keywords = {estimation, regularization}, number = {11}, pages = {2184--2195}, posted-at = {2006-04-26 07:00:52}, priority = {2}, title = {Financial markets: very noisy information processing}, url = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=726786}, volume = {86}, year = {1998} }

TY - JOUR ID - citeulike:600939 L3 - citeulike-article-id:600939 N2 - We report new results about the impact of noise on information processing with application to financial markets. These results quantify the trade-off between the amount of data and the noise level in the data. They also provide estimates for the performance of a learning system in terms of the noise level. We use these results to derive a method for detecting the change in market volatility from period to period. We successfully apply these results to the four major foreign exchange markets. The results hold for linear as well as nonlinear learning models and algorithms and for different noise models IS - 11 JF - Proceedings of the IEEE EP - 2195 TI - Financial markets: very noisy information processing VL - 86 SP - 2184 KW - estimation KW - regularization N1 - Looks at effect of noise and training size on "stable" learning systems. Expected squared error grows linearly with variance, and 1/N with N=training size AU - Magdon-Ismail, M AU - Nicholson, A AU - Abu-Mostafa, YS PY - 1998/// UR - http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=726786 ER -