A Judgment Method for Capacity Properties of Test Sub-Samples

Abstract

The invention discloses a test subsample capacity characteristic determination method. The method comprises steps that S1, original test data of a target subsample is acquired, and statistics distribution of the target subsample is determined according to the acquired original test data; S2, according to the statistics distribution determined in the step S1, target estimation precision during estimation is determined; S3, a sample capacity demand under the statistics distribution of the step S1 is calculated when target estimation precision is satisfied; and S4, capacity characteristics of the target subsample are determined according to relationships between the capacity of the target subsample and the sample capacity demand acquired in the step S3. The method is advantaged in that the method is simple to realize, quantitative test subsample capacity characteristic determination can be realized, determination efficiency and precision are high, and data analysis precision is high.

Description

technical field [0001] The invention relates to the technical field of data analysis, in particular to a method for judging capacity properties of test samples suitable for index identification and test analysis. Background technique [0002] Index identification is an important step in the process of product or system design, development or after completion. It is a process of checking whether the product or system meets the design goals. It is a key technology in various industrial fields to test the important performance of the product. . Due to the constraints of test conditions, it is difficult to achieve medium and large sample sizes (hundreds or even tens of thousands) of test data when conducting on-site tests on equipment systems with large losses, high costs, and difficult to reproduce. The capacity is small. [0003] Among the current statistical theories, different statistical theories apply differently to the traits of different sample sizes (that is, the size...

AI Technical Summary

Problems solved by technology

At present, when performing index identification or test analysis on sub-sample experiments, the classic frequency statistical method is usually directly used without considering the characteristics of the sub-sample size, that is, whether it belongs to a large sub-sample size or a small sub-sample size, and the characteristics of the sample size are determined. It is also based on empirical judgments. Only when the classic frequency statistical method cannot meet the precision requirements required for statistical inference, the analysis method for small samples is further considered, and then the analysis method for small samples is used again to analyze to meet the accuracy requirements, so that The data analysis process is complex and the accuracy of data analysis is low, and it is impossible to accurately adopt the appropriate data analysis method according to the characteristics of the sample size

[0006] When performing index identification and test data analysis on industrial equipment, the loss is usually large, the cost is high, and it is difficult to reproduce. If the appropriate data analysis method cannot be determined before relevant data analysis, different data analysis methods are used to achieve The required precision requirements will make the test loss large, greatly increasing the test cost and test complexity

Images