摘要
根据北瓦地区岩心样品的地球化学数据处理结果,介绍多元正态总体假设检验方法与对应分析相结合在该区铅锌矿化带识别中的应用。从全体数据的对应分析结果出发,将见矿钻孔的样品得分值经稳健处理后作为铅锌矿化带的已知样本,并以马氏距离为统计量对未见矿钻孔的样品进行识别,在一定信度下作出了该区已知铅锌矿化带在深部存在较有意义延伸的统计推断。
It is a basic geologial problem to distinguish an individual sample whether it belongs to a single known geological body or not. According to the geochemical data from Beiwa, Liaoning province,the method for test of hypothesis of mulitivariate normal population is used to distiguish the unknown Lead-Zinc mineralization zone. The effectiveness of this method can be improved by employing the results of the correspondence analysis as the basic data of the test of hypothesis. The unknown individual are recognizaed by testing the Mahalanobis distance statistics D = (X-X) S 1 (X-X) which based on the hypothesis of normal distribution. The inference we make is that the known mineralization zone in Beiwa may be extended down to the depth of 400 meters.
关键词
铅锌矿床
矿化
假设检验
地球化学
test of hypothesis, correspondence analysis, mineralization zone recognization, geochemistry
