logo Dunamath Outlier detector
 

Skip Navigation Links.

Skip Navigation Links. Skip Navigation Links. Skip Navigation Links. Skip Navigation Links.

 

 plot meeting

fdp gif

  
Curve 1

  
Curve 1

  

 

 

Original: English Log In Sign Up not logged
                                                                                             

Label An outlier is an observation which deviates so much from the other observations as to arouse suspicions that it was generated by a different mechanism” (Hawkins 1980).

Label Here you can identify if there are potential outliers in your data sample. The non-existence of outliers is an assumption of many statistical tests. Outliers can be caused by an error in the data collection, therefore it is always useful to look for outliers in your data and then investigate if there is really a problem with it. Typical applications are:

1 / 3
You measured the transaction time of your process and now, starting your statistical analysis, you want to know if there are potential outliers in your data.
2 / 3
As a supervisor of an administrative process, you have developed a different method to perform a task. Now you wish to confirm statistically if the average time of the new method is better. Prior to that, did you check if there are outliers in your data?

3 / 3
You measured the interarrival time at a supermarket cashier. Is there any measured value that does not belong to the same probability distribution of the process?


Label Here we perform 2 tests for outliers based on non-parametric statistical approaches for univariate data. It means it works for Normal or non-Normal distributed data.

Label Just log in/sign up and enter your data as described below.

 

Step 1) Paste your sample data in the text box.

(up to 2000 numbers separated by at least one space)

Ex: 113.47 86.62 91.99 86.56 111.51 84.44 78.73 111.06 91.53 107.23

  

Find us on facebook icon

Remember you can access this website in your mobile phone and add it to your favorites.

Help us to develop the tool. Make a donation.

All rights reserved.