Statistics tools, including a kappa calculator and a Lin's concordance calculator.
- Python 3.x
- Python dependencies (see
requirements.txtfor a complete listing) - Windows, MacOS, Linux or any other *nix OS
This project uses Python 3.x unittest module for tests. You may need to install the
requirements first. Also, if you would like to view the test coverage, please
install the coverage module using pip install coverage. For more information check the pip manual (pip -h).
To fetch the libraries that this code depends on, run this command within the statcalc folder
pip install -r requirements.txtFor running the tests.
python3 test.pyFor measuring the coverage and producing an HTML report.
coverage run tests.py
coverage htmlFor updating the coverage badge, run the commands above, and then (after installing the
coverage-badge module) run the
coverage-badge command.
coverage-badge -o coverage.svgSample usage for kappa.py:
$ python kappa.py --npp 1.0 --npa 1.0 --nap 2.0 --naa 2.0 --kappatest 0.5
Results for 2x2 Interrater table
+-----------+-----------+--------+
| Rater A | Rater B | None |
+===========+===========+========+
| | present | absent |
+-----------+-----------+--------+
| present | 1 | 1 |
+-----------+-----------+--------+
| absent | 2 | 2 |
+-----------+-----------+--------+
+-------------------------------------------------------------+
| kappahat = 0.0, (kappa+ = 0.0. kappa- = 0.0) |
+-------------------------------------------------------------+
| s.e.(0) = 0.3849, s.e.(kappahat) = 0.3849 |
+-------------------------------------------------------------+
| |
+-------------------------------------------------------------+
| Hypothesis test p-values |
+-------------------------------------------------------------+
| One-sided test, H0 is kappa =<0 |
+-------------------------------------------------------------+
| p = Prob[>kappahat, given that kappa=0] = 0.5000 |
+-------------------------------------------------------------+
| |
+-------------------------------------------------------------+
| One-sided test, H0 is kappa =< 0.5 |
+-------------------------------------------------------------+
| p = Prob[>kappahat, given that kappa= 0.5] = 0.9030 |
+-------------------------------------------------------------+
| |
+-------------------------------------------------------------+
| Two-sided test, H0 is kappa= 0.5 |
+-------------------------------------------------------------+
| p = Prob[>|kappahat-kappa|, given that kappa= 0.5] = 0.1939 |
+-------------------------------------------------------------+Sample usage for kappa_simple.py:
$ python kappa_simple.py --npp 1.0 --npa 1.0 --nap 2.0 --naa 2.0 --kappatest 0.5
Results for 2x2 Interrater table
+-----------+-----------+--------+
| Rater A | Rater B | None |
+===========+===========+========+
| | present | absent |
+-----------+-----------+--------+
| present | 1 | 1 |
+-----------+-----------+--------+
| absent | 2 | 2 |
+-----------+-----------+--------+
+------------------------------------------------------------------+
| estimated kappa = 0.0 |
+------------------------------------------------------------------+
| s.e.(0) = 0.3849, s.e.(estimated kappa) = 0.3849 |
+------------------------------------------------------------------+
| Hypothesis test p-values |
+------------------------------------------------------------------+
| One-sided test, H0 is kappa <= 0.5 |
+------------------------------------------------------------------+
| p = Prob[>estimated kappa, given that kappa= 0.90300.5] = 0.9030 |
+------------------------------------------------------------------+Try python kappa.py -h for more or python kappa_simple.py -h.
Sample usage for links_concordance.py:
$ python lins_concordance.py "1~2~2~3~3~4"
+----------------------------------------------------------+
| Concordance Results |
+==========================================================+
| Sample concordance correlation coefficient (pc) = 0.5714 |
+----------------------------------------------------------+
| Lower one-sided 95% CL for pc = -0.193 |
+----------------------------------------------------------+
| Lower two-sided 95% CL for pc = -0.343 |
+----------------------------------------------------------+
| Upper one-sided 95% CL for pc = 0.9043 |
+----------------------------------------------------------+
| Upper two-sided 95% CL for pc = 0.9298 |
+----------------------------------------------------------+Try python lins_concordance.py -h for more.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent