The Promise của Bootstrap "Một sai lầm nghiêm trọng nhiều hơn nữa xuất hiện để được tham gia vào các giả định của Galton rằng giá trị của dữ liệu, cho mục đích mà họ đã dự định, có thể được tăng lên sắp xếp lại các so sánh. Thống kê hiện đại quen thuộc với các khái niệm mà bất kỳ cơ thể hữu hạn của dữ liệu chứa chỉ một số lượng hạn chế thông tin, bất kỳ điểm nào dưới kiểm tra rằng giới hạn này là thiết lập được bản chất của chính các dữ liệu, và. | C H A P T E R 6 The Promise of the Bootstrap A much more serious fallacy appears to be involved in Gallon s assumption that the value of the data for the purpose for which they were intended could be increased by rearranging the comparisons. Modern statisticians are familiar with the notions that any finite body of data contains only a limited amount of information on any point under examination that this limit is set be the nature of the data themselves and cannot be increased by any amount of ingenuity expended in their statistical examination that the statistician s task in fact is limited to the extraction of the whole of the available information on any particular issue. If the results of an experiment as obtained are in fact irregular this evidently detracts from their value and the statistician is not elucidating but falsifying the facts who rearranges them so as to give an artificial appearance of regularity. Fisher 1966 Introductory Remarks The wisdom of Fisher s critique of Francis Galton s analysis of data from Charles Darwin s experiment on plant growth holds true today for the analysis of environmental data as it was when penned in 1935 in regard to experimental design in the biological sciences. The point is that a given set of data collected for a specific purpose contains only a limited amount of information regarding the population from which they were obtained. This limited amount of information is set by the data themselves and the manner in which they were obtained. No amount of ingenuity on the part of the data analyst can increase that amount of information. In order for the information contained in any set of data to be useful one must assume that the data at hand are representative of the entity about which we desire information. If we desire to assess the risk of an individual moving around a residential lot then we must assume that the soil samples used to assess the analyte concentration on that lot truly represent the concentrations to .
