I am hiring developers to implement new algorithms for the [url removed, login to view] Framework ([url removed, login to view]). This offer is for someone to implement all 9 methods for computing sample quantiles currently available in the GNU R software (and also described in Wikipedia at [url removed, login to view] )
The [url removed, login to view] Framework is written entirely in C#. The outcome of the work would need to be licensed under the MIT License.
The implementations can be delivered as a set of 9 static methods in a standalone .cs file. It is not necessary to incorporate the methods in the [url removed, login to view] Framework, though if you feel like posting a pull request to the project's GitHub page it would be very much appreciated.
The 9 different methods are described in:
- Wikipedia: [url removed, login to view]
- R's manual page: [url removed, login to view]
- This paper: https://www.amherst.edu/media/view/129116/original/Sample+[url removed, login to view]
Note: even though the source code for R's quantile function is available, any implementations should not have been based at all at this source code. Looking at the source of R should be regarded as expressly forbidden for the completion of this task.
Also note that, after this work is complete, there are many other similar works that you might be able to apply for.
The implementation should assume the data is given as a double array.
You should not assume the data is ordered. If needed, you can order it using any standard sorting algorithm already present in the .NET Framework.
You do not need to care about missing values (null) or non-real double values (Double.NaN, Double.PositiveInifinity, or Double.NegativeInfinity).