Modern Statistical Computing

Statistical computing is a highly sought-after analytical data analysis skill both in professional and research environments. This course presents modern graphical displays and data manipulation methods, interactive and reproducible reporting, emphasizing statistical methods and computation related to regression and classification methods such as linear, generalized linear and non-linear models. The concepts are introduced in R (https://www.rproject.org) one of the leading programming languages for statistical computing. Knowledge of R is highly-valued by companies in many sectors for positions related to data science, quantitative analysis or finance. R is also an indispensable tool in most research fields, including Economics, Finance, Marketing, Biomedicine, etc. R provides a rich set of off-theshelf data analysis tools, and the possibility to design our own data processing and analyses. R runs in all operating systems (Windows, Mac, Unix-like) and is a free open-source language that is enhanced by an extensive list of user-contributed packages. The purpose of this course is to introduce students to statistical computing, including flexible regression data analysis methods, and to advanced R skills. The idea is that students learn by doing. Therefore, there is a strong applied emphasis, all concepts are driven by examples discussed in class, where students are given the code to reproduce them. Students will become skilled in applications of elementary statistical methods, with an emphasis on data exploration, graphics and programming. Focus will also be placed on opportunities to enhance the learning experience in other statistical courses.

 

At the end of the course, students wll have learned

-to use a fundamental data analysis tool for quatitative analytical methods.

-programming, data handling, exploratory data analysis, linear / generalised linear /non -linear regression, summarising data, effective graphics, model-free computational methods (bootstrap, permutation tests, cross-validation)

-Preparing notebooks to automatically perform quantitative analyses and create reports in formats such as pdf and html, with interactive elements.

 

The course is constituted by lectures and practice with laptop computers.The teaching philosophy is that students learn by doing.

Classroom sessions are normally split into a lecture and a practice part.

 

20% Homework + Class contribution

40% Controls in-class exercises

40% Final Project

 

The Final Project (in groups of 2 students) requires a report to be submitted of up to 10 typed pages (not counting appendices). Students will select their projects from topics of their own interest (accepted by the course instructors) and will make a brief oral presentation at the end of the course.

 

Up to two (2) absences -  No penalty

Three (3) absences - 1 point subtracted from the final grade (on a 10 point scale)

Four (4) absences - 2 points subtracted from the final grade (on a 10 point scale)

Five (5) absences - The student receives an INCOMPLETE grade for the course

 

The BISS attendance policy does not distinguish between justified or unjustified absences. The student is deemed responsible to manage his/her absences.

Emergency situations (hospitalization, family emergency, etc.) will be analysed on a case-by-case basis by the Academic Director of the UPF Summer School.

 

Students are encourage to consult the following sources on their own. 

Wickam, H., Grolemund, G. R for the Data Science. O’Reilly, https://r4ds.had.co.nz.

Lander, Jared P. R for Everyone: Advanced Analysis and Graphics. Boston etc.: Pearson Education, Inc, 2017.