In statistics, an outlier is a single observation "far away" from the rest of the data.
In most samplings of data, some data points will be further away from their expected values than what is deemed reasonable. This can be due to systematic error or faults in the theory that generated the expected values. Outlier points can therefore indicate faulty data, erroneous procedures, or areas where a certain theory might not be valid. However, a small number of outliers is expected in normal distributions.
Estimators not sensitive to outliers are said to be robust.
2006-06-13 20:38:09
·
answer #1
·
answered by nick ramsey 4
·
0⤊
0⤋
An outlier is a piece of data that is much greater or less than most of the other data points.
For example, in this data set...
5,7,6,7,5,7,7,22,5,6,5
...the "22" is the outlier.
The more or less one number in a set is, the more it affects the mean. So a single big or small outlier can have a big effect on the mean.
The median is the "middle number", so one outlier will simply move the median "one number" higher or lower. An outlier may not make much difference to the median.
The mode is the "most common" number, and since outliers don't appear in a set very often (usually just once!) an outlier won't usually change a mode at all.
This should get you started. Take a look at the data set above. Take the mean, median and mode with and without the outlier.
2006-06-13 13:06:29
·
answer #2
·
answered by Polymath 5
·
0⤊
0⤋
Outliers are events or data points which are outside the expected and predicted norm.
I am a nurse, so I will answer in "hospital" terms. Suppose you go into the hospital for surgery on your knee. Most people will go home in less than 2 days. However; if someone had to stay for 3 weeks, he or she would be outside the expected length of stay, therefore considered an "outlier".
When you look at data on a chart, you will see sharp peaks above or below the average. When you see these points of data, it may prompt you to ask the question "Why?". Many times these data points can be explained as "outliers".
Of course, the next question would be "why?" In the case of the hospital stay, you might ask if there were complications in the surgery.
2006-06-13 13:18:21
·
answer #3
·
answered by tjm617 1
·
0⤊
0⤋
An outlier is a term used with averages: It is the number that might set the average off. An example might be where we have a set of numbers of the ages of people at a kids party, and we are trying to find an average: 4, 5, 5, 5, 6, and a 47-year old clown. This guy would set the average off a bit, from an average of 5 to something waaaay different. His age would be an outlier.
2006-06-13 12:52:19
·
answer #4
·
answered by buster d 2
·
0⤊
0⤋
An outlier is a piece of data that doean't fit with the reat of your data. If you were doing an experiment of how long a dog lives and you looked at 100 dogs. lets say that 99 of the dogs lived 5-10 years, but one dog only lived one year...it would be an outlier. the measures of central tendency can be shifted to the left with an un usually low outlier, or to the right with an unusually high outlier. In the example that I gve you, the central tendency is reduced, or moved to the left.
2006-06-13 12:58:36
·
answer #5
·
answered by jtiernan07 2
·
0⤊
0⤋
an outlier is the number that is either larger or smaller than the other numbers in the set of data.
ex. 1,2,3,4,5,100
the outlier would be 100 because the other numbers are closer together.
2006-06-13 12:55:25
·
answer #6
·
answered by foxy 4
·
0⤊
0⤋
an out liner is an unually unfitting number in an exquation. like if you have 1 2 3 4 456, 456 would be an outliner. outliners are the extremes in you date and really don't match any number in the set.
2006-06-13 12:53:42
·
answer #7
·
answered by Anonymous
·
0⤊
0⤋