You will need
- - several measurement results or other sample;
Measure at least 3-5 times to be able to calculate the actual value of the parameter. Total the results and divide them by the number of measurements you got a valid value, which is used in the problems is true (it is impossible to determine). For example, if the measurement gave the result 8, 9, 8, 7, 10, the actual value will be (8+9+8+7+10)/5=8,4.
Find the absolute error of each measurement. From this measurement subtract the actual value, the signs of neglect. You will receive 5 of the absolute error, one for each dimension. In the example they are equal 8-8,4 = 0,4, 9-8,4 =0,6, 8-8,4=0,4, 7-8,4 =1,4, 10-8,4=1,6 (taken modules results).
To find the relative error of each measurement, divide the absolute error to the actual (true) value. Then multiply the result by 100%, it's usually at a percentage measured by this value. In the example locate the relative error is therefore: δ1=0.4 a/8,4=0,048 (or 4.8%), δ2=0,6/8,4=of 0.071 (or 7.1 %), δ3=0.4 a/8,4=0,048 (or 4.8%), δ4=1,4/8,4=0,167 (or 16.7%), δ5=1,6/8,4=to 0.19 (or 19%).
In practice, for the most accurate display errors using the standard deviation. To find it, erected in the square of the absolute error measure and add together. Then divide this number by (N-1), where N is the number of measurements. Calculating the root of the result, you will get a standard deviation that characterizes the error of measurement.
To find the limit of the absolute error, find the minimum number obviously exceeding the absolute error or equal to it. In the example just select the highest value of 1.6. It is also sometimes necessary to find the maximum relative error, in this case, find the number that is greater than or equal to the relative error in the example it is equal to 19%.