--- title: "Poisson Pseudo-Maximum Likelihood (PPML) Model with Cluster-Robust Standard Errors" output: rmarkdown::html_vignette bibliography: "references.bib" vignette: > %\VignetteIndexEntry{Poisson Pseudo-Maximum Likelihood (PPML) Model with Cluster-Robust Standard Errors} %\VignetteEngine{knitr::rmarkdown} %\VignetteEncoding{UTF-8} --- ```{r, include = FALSE} knitr::opts_chunk$set( collapse = TRUE, comment = "#>" ) ``` We will estimate a Poisson Pseudo-Maximum Likelihood (PPML) model using the data available in this package with the idea of replicating the PPML results from Table 3 in @yotov2016advanced. This requires to include exporter-time and importer-time fixed effects, and to cluster the standard errors by exporter-importer pairs. The PPML especification corresponds to: \begin{align} X_{ij,t} =& \:\exp\left[\beta_1 \log(DIST)_{i,j} + \beta_2 CNTG_{i,j} +\right.\\ \text{ }& \:\left.\beta_3 LANG_{i,j} + \beta_4 CLNY_{i,j} + \pi_{i,t} + \chi_{i,t}\right] \times \varepsilon_{ij,t}. \end{align} We use `dplyr` to obtain the log of the distance. This model excludes domestic flows, therefore we need to subset the data also with `dplyr`. Required packages: ```{r} library(capybara) ``` We can use the `fepoisson()` function to obtain the estimated coefficients and we add the fixed effects as `| exp_year + imp_year` in the formula. Model estimation: ```{r} fit <- fepoisson( trade ~ log_dist + cntg + lang + clny + rta | exp_year + imp_year, data = trade_panel ) summary(fit) ``` The coefficients are almost identical to those in Table 3 from @yotov2016advanced that were obtained with Stata. The difference is attributed to the different fitting algorithms used by the software. Capybara uses the demeaning algorithm proposed by @stammann2018fast. ```{r} fit <- fepoisson( trade ~ log_dist + cntg + lang + clny + rta | exp_year + imp_year | pair, data = trade_panel ) summary(fit, type = "clustered") ``` The result is similar and the numerical difference comes fom the variance-covariance matrix estimation method. Capybara clustering algorithm is based on @cameron2011robust. # References