All you need is love and R
Meetup Description
Michela Cameletti from the University of Bergamo talks about academic (teaching and research) activities for which R is used.
About Speaker
Michela Cameletti is an Associate Professor in Statistics at the Department of Economics, University of Bergamo. Previously, from 2008 to 2018, she was Assistant Professor in Statistics at University of Bergamo. In 2001 she graduated in Statistics, Demography and Social Sciences (University of Milan-Bicocca, Italy) and obtained in 2007 her PhD in Statistics from the same University. In 2007 she was research fellow at the Statistical and Applied Mathematical Sciences Institute (NC, USA), in 2011 visiting researcher at NTNU University (Trondheim, Norway), and in 2012 honorary research fellow at the MRC Centre for Environment and Health Department of Epidemiology and Biostatistics, Imperial College London (UK). She leaded the local unity of Bergamo for 2 national projects funded by MIUR (PRIN2015 “Environmental processes and human activities: capturing their interactions via statistical methods” and FIRB 2012 “Statistical modeling of environmental phenomena: pollution, meteorology, health and their interactions”).
Contact Speaker
All you need is love…..and R
As an academician, Michela has used R in many ways.
The packages she uses the most
Data manipulation and plotting:
standard `R` code and objects
`tidyverse` collection
Spatial objects manipulation and mapping:
- `rgdal`, `spdep`, `sp`, `sf`, `lattice`, `fields`, `maptools`, `leaflet`, …
Modelling:
- `R-INLA`, `rpart`, …
Parallel computing:
- `parallel`, `snow`, …
a) Interactive research communication
Some of her interactive research communication includes her Shiny app link from her research paper A spatio-temporal analysis of NO2 concentrations during the Italian 2020 COVID-19 lockdown.
b) Interactive research communication
Her other work on Regional excess mortality during the 2020 COVID-19 pandemic: a study of five European countries, uses data for 2015-2019 to apply Bayesian spatio-temporal models to quantify the expected weekly deaths at the regional level had the pandemic not occurred in England, Greece, Italy, Spain, and Switzerland. The shiny app estimates the excess mortality caused by the COVID-19 pandemic during 2020 in five European countries: England, Greece, Italy, Spain and Switzerland.
Lab notes example
The notes posted online for her students can be accessed anywhere there is internet connectivity. See the labs’ notes for the Machine Learning for Economics. The course notes were created using Bookdown. An open-source R package�that facilitates writing books and long-form articles/reports with R Markdown.
Stem package
Unfortunately, the Stem package that was developed by Michela has been removed from CRAN. However, the last documentation can be seen at R documentation.
This package focused on spatio-temporal hierarchical models. The package includes functions for maximum likelihood estimation (based on Kalman filtering and EM algorithm), for computing the parameter standard errors (using a parametric spatio-temporal bootstrap) and for spatial mapping
YouTube link
Watch the full recording of the meetup session here.