关键词: 投影寻踪自回归, 地震时间序列, 视二维预测

Abstract: Projection pursuit (PP) is a new statistic method, which is good at analyzing non- normal.and non- -linear high-dimensional data. It searches for the project direction reflecting on the structurecharacteristics of high-di. mensional dada objecti vely by projecting and reducing di mensions, and solves" dimension curse" and non nomality and non-linearity among high- dimensions data. The article com -bines the PP technique with auto- regression model of time sequence analysis, and builds up the predic-tion model of projection pursuit au to- regression(PPAR). PPAR model tries to realize tw o dimensionalforecast of magnitude and time, i. e. forecasting the magnitude and time of an event in the fixed re-search region, and creates the pojection pursuit auto- regression model of tw o- dimensional seismic timesequence. In the study, we choose finst the northern Tïanshan area as the test site, and the results ofthe regression fitting and pretest test are good, so we could realize tow - -dimensional forecast. Consider-ing the value of forecast practice, i. e. moderately strong earthquake, we take the whole Tianshanmountain area as our research area. Let the magnitude th esholds of time sequence are 5.0 and 5.5 re-spectively, and build up the models with data of unde leted aftershocks and deleted- -aftershocks. Com-paring the two models, the latter is better than the former, particularly is to forecast time sequence.Their qualified ration of pretest tests are both high, so they are available for forecasting magnitude andtime of an event.

Key words: Project pursuit auto-regression, Seismic time sequence, Apparent two-dimensional forecast