Unit STATISTICAL METHODS FOR FINANCE
- Course
- Finance and quantitative methods for economics
- Study-unit Code
- A000204
- Location
- PERUGIA
- Curriculum
- Finanza ed assicurazione
- Teacher
- Luca Scrucca
- Teachers
-
- Luca Scrucca
- Silvia Pandolfi (Codocenza)
- Hours
- 21 ore - Luca Scrucca
- 21 ore (Codocenza) - Silvia Pandolfi
- CFU
- 6
- Course Regulation
- Coorte 2020
- Offered
- 2020/21
- Learning activities
- Caratterizzante
- Area
- Matematico, statistico, informatico
- Academic discipline
- SECS-S/01
- Type of study-unit
- Obbligatorio (Required)
- Type of learning activities
- Attività formativa monodisciplinare
- Language of instruction
- English
- Contents
- Introduction to statistical software R. Statistical inference: method of maximum likelihood estimation. Regression: simple and multiple linear regression model; logistic regression. Analyses and real data case studies.
- Reference texts
- Weisberg, S. (2014) Applied Linear Regression, 4th edition, Wiley
Ruppert, D. and Matteson, D.S. (2015) Statistics and Data Analysis for Financial Engineering, 2nd ed., Springer
Supplemental material will be provided by the instructor during the course. - Educational objectives
- Upon completion of this course the student will be able to apply to real data problems the methods of statistical inference, linear and logistic regression, studied through the use of the software R.
- Prerequisites
- Fundamentals of statistics (descriptive statistics and statistical inference).
- Teaching methods
- Lectures and use of statistical software R.
- Other information
- Attending classes is not mandatory but strongly advised.
- Learning verification modality
- The exam consists in a written test with open questions about theoretical topics and interpretation of the output from the R software for data analysis.
- Extended program
- Introduction to the statistical software R.
The method of maximum likelihood: likelihood and log-likelihood function, maximum likelihood estimates (MLEs), Score function and Fisher information, statistical properties of MLEs, numerical computation of MLEs, likelihood ratio test and likelihood-based confidence intervals. Fitting distributions to log-returns by maximum likelihood.
Regression models: simple and multiple linear regression, parameter estimation, least squares method and maximum likelihood method, hypothesis testing, confidence intervals, categorical predictors, analysis of variance, goodness of fit, transformations, model selection methods, diagnostic and residual analysis, multicollinearity. Logistic regression: model specification, parameter interpretation and estimation.