# 87497 - STATISTICS APPLIED TO FINANCIAL AND INSURANCE RISK MANAGEMENT

## Learning outcomes

At the end of the course, students should be able to interpret, understand and explain the results of applications of univariate and multivariate statistical methods, applied to analyses of financial, banking and insurance risk. Methods for the analysis of cross-sectional, time-series, and panel (i.e., longitudinal) datasets will be explored, including: the appropriate graphical presentations of different types of data, univariate and multivariate comparisons of variation using summary statistics, and regression techniques for the analysis of different types (e.g., continuous, categorical) of data. Students will be able to describe the results of statistical analyses in words that are intelligible to their (non-statistician) peers and policy-makers.

## Course contents

1. Introduction: Principles of Risk and Insurance; Portfolio Theory.

2. Questions, Theories, Hypothesis Testing

3. Measurement and Summarization

4. Statistical Inference

5. Regression Analysis

6. Multiple Regression Analysis

7. Statistical Analysis of Time-Series and Panel Datasets

8. Applied Statistical Studies and the Law

*Epstein L and Martin AD (2015) An Introduction to Empirical Legal Research, Oxford University Press: Oxford; Chapters 1-3 and 6-10 (inclusive).

*Cummins JD (1991) Statistical and Financial Models of Insurance Pricing and the Insurance Firm, Journal of Risk and Insurance, 58(2): 261-302.

## Teaching methods

The course includes lectures and essential readings (marked with an asterisk (*), above) to introduce the key concepts, supplemented by the textbook chapter readings and supplementary reading in the form of journal papers.

Apart from the first lecture, each subsequent lecture is accompanied by statistical laboratory work. The laboratory work requires class members to analyse and interpret real datasets (which will be supplied for this purpose) using specialist statistical software (i.e., Stata), with the support of Professor Connelly who will demonstrate these methods in the statistical laboratory and in lectures.

In addition to the required readings, a range of additional resources (e.g., links to resources on the web) will also be supplied to assist class members, especially for further guidance in the use of the specialist statistical software.

Note: The emphasis of this course is applied data analysis, on understanding the results of statistical analysis, and on presenting the results in a meaningful way to an educated audience (of non-statisticians). Students will be required to analyse and present data and to explain the results, in words. (Students will not be required, for example, produce formulae, theorems, etc. or to conduct analyses using statistical software under exam conditions.)

## Assessment methods

After the fourth laboratory session, students will be required to (i) submit a log of at least three of their laboratory work sessions (10% of the course weight); and (ii) to submit a short (<5-page) written analysis of at least one empirical/applied project (30% of the course weight) before the final examination; and to sit a final, open-book, examination of 2 hours’ duration (60% of the course weight) at a date to be fixed.

Examples of the type of log that is required for (i) and the type of report that is required for (ii) will be supplied to the class in the first week of lectures.

For the final examination, students are permitted to bring the required textbook to the exam room, should they wish to do so, and to rely upon any of its contents during the examination. One double-sided A4 page of notes (handwritten or typed) may also be brought into the examination. Calculators and other electronic devices are neither required nor permitted for the final examination.