diff --git a/05-multidimensional.Rmd b/05-multidimensional.Rmd new file mode 100644 index 0000000..f5e4c01 --- /dev/null +++ b/05-multidimensional.Rmd @@ -0,0 +1,11 @@ +--- +output: + pdf_document: default + html_document: default +--- + + + +# Multidimensional Poverty + +## Example Calculation diff --git a/05-future.Rmd b/06-covariance.Rmd similarity index 91% rename from 05-future.Rmd rename to 06-covariance.Rmd index d495fff..8a4bb91 100644 --- a/05-future.Rmd +++ b/06-covariance.Rmd @@ -5,14 +5,9 @@ output: --- +# Covariance Matrix -# Future Research - -## Multidimensional Poverty - -### Example Calculation - -## Covariance Matrix +The current `convey` software does not account for the covariance. The example below describes this calculation and presents what is, in most cases, a limited impact on the final error terms. We acknowledge that, under ideal circumstances, the covariance would be properly accounted for in the software. Modifying our software to account for the covariance remains a future goal. Influence functions and "resampling" replicates can be used to improve the inference about differences and changes between estimates. Put simply, the variance of the difference between @@ -70,7 +65,7 @@ A similar idea can be implemented with other estimators, such as inequality and However, this has not yet been implemented for linearization/influence function methods. In the next section, we show an example with the Gini index. -### Example Calculation +## Example Calculation We start by showing what has been implemented --- the resampling based approach: diff --git a/docs/1.1-install.html b/docs/1.1-install.html index 0683f24..aa3b715 100644 --- a/docs/1.1-install.html +++ b/docs/1.1-install.html @@ -13,7 +13,7 @@ - + + + + + +