Statistics is a must-needed set of tools in experimental sciences. In almost every study, methods from statistics are necessary and their justification is required for a good level publication. Despite the availability of a number of powerful and sophisticated software, it is not always obvious what analysis approach is needed in any particular case if one does not have an introductory level background in the matter.

The courses offered here are aimed to fill the gap. The courses are complementary and partialy overlapping, with the idea of addressing the needs of different people.

• Introduction to Probability Theory for Applied Statistics (remote course)
• This ten-hour course will cover the main aspects of probability theory needed to have a good command of the main tools used in applied statistics. Course topics include a description of the most common univariate and bivariate probability distributions, the distributions related to the normal distribution and a discussion of conditional probabilities. At the end of the course you will be able to understand most probability calculations in scientific papers and make your own calculations, if needed. You will know where the chi-square, the t and the F distributions come from and how they are formulated.
• This course is suitable for people with some elementary knowledge of calculus (integration and derivation) and linear algebra.
• At the end and during the course you will have access to the lecture notes and to unique and dedicated python scripts useful to understand and solve the exercises.
• Applied Statistics, Part 1 (remote course)
• This ten-hour course will cover methods for estimating means and proportions by defining and computing confidence intervals. We will then discuss methods for hypothesis testing for means and proportions using rejection region, p-value and confidence interval approaches. The tools discussed in the class will be the most used ones in everyday work with experimental data (including z-test, t-test). The course won't refer to any particular software but people used to work with any given software are welcome to use it in order to solve the exercises.
• In this course, some emphasis will be put on mathematical derivations in order to understand why and how the estimators follow certain distributions. An elementary background in calculus is very helpful. Knowledge of Probability Theory will definitively help.
• At the end and during the course you will have access to the lecture notes and to unique and dedicated python scripts useful to understand and solve the exercises.
• Applied Statistics, Part 2 (remote course)
• This ten-hour course will cover methods of single linear regression, with a derivation of the properties of the estimators of slope and intercept, including the interval estimation of the prediciton. It will furthemore cover goodness-of-fit tests and the analysis of contingency tables. It will provide an introduction to the ANOVA methods and to non-parametric mathods commonly used in statistics.
• In this course, some emphasis will be put on mathematical derivations in order to understand why and how the estimators follow certain distributions. An elementary background in calculus is very helpful. Knowledge of Probability Theory will definitively help.
• At the end and during the course you will have access to the lecture notes and to unique and dedicated python scripts useful to understand and solve the exercises.
• Introduction to Applied Statistics (remote course)
• This ten-hour course will cover the main methods of interval estimation for means and proportions, sample size estimations, simple linear regression, contingency tables and ANOVA methods. The emphasis will be on the applied side, with illustrative exercises and discussion of the methods without entering in the details of their mathematical justification.
• Knowledge of calculus and probability theory is of advantage but not necessary to follow the course.
• At the end and during the course you will have access to the lecture notes with the computations of the examples. The examples will be designed so that you can reproduce the results using you own reference software.
• Introduction to Python for Applied Statistics (remote course)
• This ten-hour course will cover the main methods of applied statistics while working together on taylored python scripts to solve the exercises. At the end of the course you will have your own Jupyter notebooks that perform the analysis of your choice, including interval estimations for means and proportions, hypothesis testing for means and proportions, sample size calculations, simple linear regression, contingency table and ANOVA.
• Some practice in programming with Python is necessary for a smooth participation. Elementary knowledge of applied statistics is a pre-requisite to participate.
• At the end and during the course you will receive the Jupyter notebooks derived during the lecture together with the lecture notes.

Each of these courses could be delivered in compact form (two days per course) or as traditional lecture series with weekly appointments. If you would like to book one of these courses for a closed group of people, please contact me at angelo.valleriani@mpikg.mpg.de

Everyone interested is welcome to join and attend the courses. At present there are no dates fixed for the courses listed above. If you intend to take anyone of these courses or you want to be informed via email about changes in the schedule, please register.