Exercise with R

These notes are a companion material to the course 'The R language and environment for statistical computing' held by the author at Università Cattolica del Sacro Cuore, Bachelor Degree in 'Economics and management'. The aim of the course is to make participants able to implement statistical procedures in order to properly solve business and economic problems by means of the open source software R (R Core Team, 2021). The reader is assumed to possess a basic understanding of statistics with regard to data analysis, probability and inference, at the level of the combined courses 'Statistica (Analisi dei dati e probabilità)'/Statistics and 'Statistica applicata'/'Applied statistics and big data' teached at Università Cattolica del Sacro Cuore. Notes are organized as follows. Sections from 1 to 3 consider the basics of R and deal with problems of descriptive statistics, according to the methodological presentation available e.g. in Boari, Cantaluppi (2018). Section 4 considers the generation of pseudo-random numbers, introduces some simulation problems and gives the basics on how to work with parallel architectures. Section 5 explains how to read/import data in R by also considering social network and open data sources. Section 6 illustrates how to solve some basic mathematical problems. In particular, graphical representations of functions in one and two variables, optimization problems and solution to system of equations and system of inequalities are considered. Section 7 proposes a short analysis of the survey on households wealth administered by the Bank of Italy (2014). Section 8 describes how to solve some basic inferential problems (fitting distributions, testing population proportions, testing mean and variances in one and two sample instances, fitting linear and generalized linear (logit) models). Finally, Appendix A gives a brief introduction to packages dplyr, ggplot2 and ggmap that are frequently used in data science applications. Sections from 1 to 7 are accompanied by exercises. Verzani (2002), Wickman, Grolemund (2016) and Williams (2017) are useful references for further methods/exercises/problems. Companion material to these notes (files, data sources and codes) can be requested to the author These notes were produced with R (version 4.0.4), knitr 1.31 and MiKTeX 21.2 on a machine with Windows 10Pro run on an Intel(R) Core(TM) i9-10900K CPU with 32GB RAM.