Shikhar Mishra
Shikhar Mishra
In R `survey`, population proportions and counts are calculated using `mean` and `total` respectively. Previously in Survey.jl we followed this approach, which can be seen in `AnalyticalSolution` branch. I am...
Currently very little testing for `ratio` method in `ratio.jl`. - [ ] Add more variety of tests for `ratio` method - [ ] Run and check with analogous commands in...
I noticed negligible standard errors (almost 0) in many domain estimation cases. like here ```julia julia> mean(:api00, :cname, bclus1) 11×3 DataFrame Row │ cname mean SE │ String15 Float64 Float64...
To integrate `_ci` function with `total`. See #190 for exemplar integration with `mean`, and keep consistency. - [ ] `total` with confidence interval options - [ ] Add docstrings -...
Post-stratification and calibration methods might take some time to be integrated into `Survey.jl`. Imputation based method(s) can be investigated and integrated more easily into the package. There is choice of...
- [ ] Docstrings for `by` above function - [x] Testing in test/mean.jl and test/total.jl - [x] Documentation usage in some of the examples and markdown files
R uses `svymean` for calculating population proportions. `svyciprop` can be used when proportions are closer to 0 and 1. See [here](https://stackoverflow.com/questions/41426055/how-to-calculate-proportion-using-survey-package), [here](https://r-survey.r-forge.r-project.org/survey/html/svyciprop.html) for more inspirations.
So far, most of the testing suite is limited to the API dataset. I suggest to improve testing by using other publicly available survey datasets. R Lumley survey textbook examples...
Based on discussion about JKn taking some time to think and implement, and the package currently having a JK1 working for the designs it is meant to work for (simple,...
See R survey [line 1832 and 1915](https://github.com/cran/survey/blob/master/R/surveyrep.R) for source code for their postratification and raking method for replicate designs.