1 / 4

# Random Errors - PowerPoint PPT Presentation

Random Errors. Suppose that I make N measurements of a certain quantity x and my measurement errors are random. Then I would report my final answer as as:. average. standard deviation. Where:.

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.

## PowerPoint Slideshow about ' Random Errors' - parley

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

Suppose that I make N measurements of a certain quantity x and mymeasurement errors are random. Then I would report my final answeras as:

average

standard deviation

Where:

As an example, suppose we measured the period of a pendulum over 5 cyclesfor a given length and we made 5 such measurements. Below I show the average and standard deviation for the measurement of the period :

So I would report my measurement of the period for this specificlength as 2.07 ± 0.02 s.

You can use Excel to find the mean (average) and standard deviationfor numbers in a row (say B2 to B6). The functions are:

AVERAGE(B2:B6)

STDEV(B2:B6)

Suppose a given set of measurements is indeed random and the setis characterized by a certain average or mean: m and a certain standard deviation: s. We assume that the distribution of measurementsfor x will follow a Gaussian distribution given by:

The constant in front of the exponential guarantees that the integralof f(x) from minus to plus infinity is 1; that is - the probability of gettingsome value is 100%. This function allows us to estimate the probabilitythat another measurement of x will deviate from the meanby somespecified amount.

The integral under the Gaussian distributionfrom (m-s) to (m+s) is the probability thatanother measurement will fall within 1 s of themean. According to the table below that is (100-31.7)% = 68.3%. Similarly, the probabilitythat a measurement is within 2 s of the meanis 95%.