标准差（国外英文资料）.docVIP

标准差 The standard deviation is the result of variance root (i.e. arithmetic square root of variance) Suppose the average of this set of data is m The variance formula s^2=1/n[(x1-m) ^2+ (x2-m) ^2+ +... (xn-m) ^2] Variance or standard deviation is that the size of the index fluctuation of a set of data, the standard deviation is the arithmetic square root of the variance, the variance or standard deviation can judge the stability of a set of data: the variance or standard deviation is greater, more unstable data; Averages can reflect the average level of a set of data; The mode is the highest number of number of a set of data, which can reflect the most mode level of a data set; A median is the average number of bits (odd numbers of data) in a set of data (or odd numbers of data), or the average number of two numbers in the middle (even numbers of data), so the median reflects the level of the middle position of a set of data Standard deviation is the most common form of quantification that reflects the degree of dispersion of a set of data, such as concrete strength, test values, and is the most important indicator of precision. Since the detection method is always error, the detection value is not its true value. But the true value is unknown. In order to ensure the accuracy and reliability of each batch of experimental results and to quantify the accuracy of the detection method, the concept of evaluation test dispersion must be introduced. Although it is impossible to know the true value of the sample, each sample always has a true value, no matter how much it actually is. As you can imagine, a good detection method, the base detection value should be very closely scattered around the real value. How not close, that will be true from the value of the big, accuracy is not good, it is impossible to imagine the discrete method, will detect accurate results. Therefore, the discrete degree is the most important and basic index of the evaluation method. The standard dev