Análisis estadístico de la elección presidencial de 2006: ？Fraude o errores aleatorios?

this paper tests some of the main fraud allegations of the 2006 presidential election in mexico with the evidence from a statistical analysis of precinct-level data. first, i analyze the data flow from the quick count, the preliminary results (prep) and the official district tally (cómputo distrital). secondly, i analyze the size and distribution of errors in the polling-station acts and the partial recount. lastly, i consider atypical precincts and the role of party representatives. my main conclusion is that, since most fraud allegations are not supported by the evidence, the presidential election outcome is statistically reliable.