Master #6
Open
Master #6
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
First commit of a routine to calculate the moment of a distribution.
The first commit of routines for both skewness and kurtosis. This needs unit testing and proper documentation.
This is the first commit of a routine to calculate the standard error for a sample. This needs testing.
This routine is to calculate the coefficient of variation. This needs testing.
This is the first commit of 9 functions to calculate quantiles. All work was taken from the Hyndman and Fan paper (1996) and a PDF of the paper can be accessed at https://www.amherst.edu/media/view/129116/original/Sample+Quantiles.pdf
All routines were sorting in the wrong order and thus producing the incorrect quantile. The first three quantile functions required some changes in terms of how g was defined (to stop compiler warnings) and for correctness.
I’ve added comments about the background of each function. A link is provided to the relevant page of R documentation which has blatantly been copied here. The routines all produce the same results to R with a test set of 50 data. This needs further testing to ensure accuracy but seems to be reasonably close so far.
I put in a generic function caller which allows users to call any of the quantile functions using the 3rd parameter (qtype). By default, quantile 7 is selected which is the same model as used by R and S. Quantile 8 is the one recommended by Hyndman and Fan (1996).
First commit of geometric mean function
First commit of harmonic mean function. Needs documentation to be added.
First commit of effect sizes functions. There are two main functions available: 1. effectSizeControl - to be used when a condition is compared against a control condition 2. effectSize - to be used when two experimental conditions are compared (i.e., neither is a control condition).
1. Adjusted spacing to 2 spaces (was 4) 2. Adjusted the routine to check the array being sent has content
|
Hi @salmoni, thanks for the changes. Sorry, I am too slow, still working on the previous functions that you submitted. I have added I will let you know when I finish with the first bunch. Thanks! |
|
Hi Evgenii,
No worries. I've hit a busy spot right now – I'm searching for my next
research contract and haven't had much time lately, but I'm hoping to get a
few hours today or tomorrow. I'm hoping to put some tests together to prove
the code. :-)
All the best,
Alan
…On 14 January 2017 at 00:13, Evgenii Neumerzhitckii < ***@***.***> wrote:
Hi @salmoni <https://github.com/salmoni>, thanks for the changes. Sorry,
I am too slow, still working on the previous functions that you submitted.
I have added skewnessA, skewnessB and centralMoment so far in this
branch: https://github.com/evgenyneu/SigmaSwiftStatistics/tree/
salmoni-master
I will let you know when I finish with the first bunch. Thanks!
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#6 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAnVun_NHDHsMp8vin7zTrykGJQIF_Lcks5rSBMPgaJpZM4LXrUc>
.
--
Dr. Alan James Salmoni Ph.D.
UX Design & Research Consultant
Web: http://www.thoughtintodesign.com
Registered in England and Wales number 7367421
VAT Number: 214 4181 42
Cellphone: +44 07428 172487
LinkedIn: http://uk.linkedin.com/in/alanjamessalmoni
Have to do statistics but don't feel confident? Try Salstat the friendly
statistics program – free and open source at http://www.salstat.com
|
Support functions have been made ‘internal static’ functions.
A short routine to extract all the unique values that occur in an array.
The first commit of a routine that extracts the unique values that occur in an array and returns them along with the frequency that each value occurs. This is useful in ranking and nonparametric statistics.
This routine ranks a vector. Tied ranks can be given the mean, minimum, maximum, first or last ranks. See the description of ranking for R for more details. https://stat.ethz.ch/R-manual/R-devel/library/base/html/rank.html
First commit
Unit tests for the quantiles routines. These probably need more work.
Function names were changed so as not to begin with an upper case letter. Functions were also made public static functions.
A function to calculate the mode of an array. This identifies not just the maximum value but also returns the indices where this value occurs.
Probably more a glitch or my exploring Github that’s doing this…
Or me - unsure which
Added some comments to explain why these variables are defined
Changed unit test names to make them more descriptive of what types of things they test
|
Hi @salmoni. I have finished working on the first batch of functions that you submitted, pushed to master and released a new version. Please let me know if you find any typos or other problems with those functions.
I have not looked at the other commits that you pushed to this branch yet, starting with "generic function caller", "geometric mean" etc. Since I made a lot of changes to your code, could you please create a new pull request on top of master with those additions? It would be easier if you could create several pull requests: a separate pull request for each new function, if that's possible. This way it will be faster for me to release the functions, one by one. Thank you so much for your contribution! I have added your name to the readme if you don't mind. |
Unit test code should be improved. Proper failures are thrown when the tests fail and the code is clearer than previously.
The first commit of some unit tests for using the uniqueValues function.
Sloppy coding first time around. I moved the declarations to a place where they are not declared if an optional ‘nil’ value is returned
The first commit for some unit tests for the frequencies function. Included are: * Empty array * Array with a single (negative) element * Array with one value multiple times * All positive elements * All negative elements * Both positive and negative elements There are many other use cases I’ve not thought of: Contributions are welcome!
Removed all ‘var’ declarations and replaced with ‘let’ reducing code length and increasing clarity.
Removed all ‘var’ declarations and replaced with ‘let’ reducing code length and increasing clarity.
This is the first commit for unit tests for the moment function. Currently, only a single data set is analysed for moments 0 - 4 inclusive, and a test for an empty array, and an array with a single value. Test results were obtained using SciPy (stats.moment) as a reference.
This is the first commit of the unit tests for the skewness and kurtosis functions. The first 2 test using a normal array for both functions. The next two test the functions with an empty array (do they return ‘nil’ correctly?) and the third two analyse an array with a single element. They probably need extending with other data sets.
This is the first commit of the unit tests for the geometric and harmonic means. These tests only test using a fairly normal array of doubles, an empty array, and an array with a single element. scipy.stats.gmean and scipy.stats.hmean were used for reference results.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hi Evgenii,
Here’s my first attempt at a pull request.
I’ve written routines for:
All the best,
Alan