search

worldbank / PIP-Methodology

Methodology page for the Poverty and Inequality Platform.
https://datanalytics.worldbank.org/PIP-Methodology/
MIT License
2 stars 1 forks source link

zeros and references #16

Closed randrescastaneda closed 3 years ago

randrescastaneda commented 3 years ago

Hi @danielmahler ,

I just added a brief explanation of how we treat the zeros in the Watts and MLD and solved the issue of the double references in the fourth chapter. The solution was to add a final chapter that includes only the references of the whole book. A few things,

  1. The explanation I provided for the Watts and the MLD is unjustified. I merely say that for consistency with previous releases or for consistency with other measures, we opted for the current decisions. Yet, I am not sure this either accurate or correct to put in a text like this.
  2. I said that we are currently doing some research on this problem of zeros, but the it might seem an exaggeration, though I know that Christoph would like to invest some time on this. I also have done some simulations that can show to you, which are about how much the mld changes depending on the size of the share of zeros in the sample.
  3. I have set up my Rstudio to break any line longer than 70 something lines in .Rmd documents and I did not realized about that after I had made my modifications. It may seem to you that I modified all the line of the document, but it is just the break of the lines. Nothing more.

I think is all for now, but please let me know if you have any question.

Best,