Financial Data Analysis

FIE401A Financial Data Analysis

  • Topics

    Topics

    The course teaches  econometrics as applied to financial markets and firms using lectures and  cases. It corresponds closely to FIE401B and covers the same econometric topics as other methods courses, such as ECN402. In addition, it looks into topics that students of Finance and Macroeconomics should master, such as event studies or more advanced time series models. The course specifically covers applications in Finance and Macroeconomics.

     

    Topics:

    • Introduction to Statistical Software
    • Elements of Statistics
    • Simple and Multiple Regression Models; Omitted Variable bias
      • Potential Cases: CAPM beta estimation and Fama-French three factor asset pricing models.
    • Functional Form: Multicollinearity, Dummy Variables and Model Selection
      • Potential Case: Store Profitability Analysis
    • Regression with a Binary Dependent Variable
      • Potential Cases: Bank Profitability and Customer Retention, Credit Scoring
    • Regression with Panel Data
      • Potential Case: Compensation of star employees
    • Instrumental Variables Regression and Causal Analysis
      • Potential Case: Commodity Pricing
    • Time Series Analysis: Random Walks, Forecasting and Co-intergation
      • Potential Cases: Real Estate Pricing; Currency issues
    • Event Studies
      • Potential Case: Computing Damages from stock market information for litigation
    • Hedonic Pricing (if time permits)
      • Potential Case: Pricing of tankers using observed market prices.

     

    Each topic is initially introduced in a lecture that covers the necessary (formal) background. For example, we derive the OLS estimator in the first lecture and we discuss the assumptions behind it. We then use a case to practice how to use this technique. When we discuss binary depend variables, we hear about Maximum Likelihood Estimation. 

    Please note that the cases can change over time and are only an indication of a potential application of each technique learned.

    Why do we do what we do and why do we think cases are a good way of teaching you econometrics? Statistical software like R or Stata have made it very easy to do an actual statistical analysis. The real problem is to be able not only run the final regression but to be able to do the whole data analysis from formulating the research question, the translation of the research question into a hypothesis, selecting the data, transforming the data into a dataset and the eventual analysis. Most econometrics courses will prepare the final data for you and do not expose you to all these other steps. Yet in real life these steps matter, often more than the actual analysis itself. Cases are the best way to teach you this.

    To put this differently, finance professionals typically have access to a lot of data about their business and their industry. These data pertain to financial data, but also inputs, outputs, pricing, sales volumes, customer base and many other characteristics. The volume of information that is readily available is always increasing as computers reach into every corner of the business and become more integrated across the business. How can managers use these data for sound financial decision making?

    Our course considers some of the core issues that arise in finance and macroeconomics, such as estimating company betas, running an event study to estimate the effect of corporate events on stock prices and firm value, analysing mutual fund performance, analysing bank profitability, testing asset price series for random walks and searching for arbitrage opportunities, testing for long-run relationships between random walks, and predicting default rates on mortgage loans.

    We specifically think of our course as a "Small Data" course, as it teaches students how to analyse typical data that firms routinely collect or data that is readily available from financial data providers. It provides a good starting point for any big data or machine learning course, as a large part of core techniques in this fields, OLS and Logistics regression, are covered extensively in the course.  

  • Learning outcome

    Learning outcome

    The course gives a broad survey of the use of econometrics as applied to financial markets and firms.

    Knowledge - The candidate...

    • has advanced knowledge within financial econometrics and its applications;
    • has thorough knowledge of the scholarly theories and methods in financial econometrics;
    • can apply this knowledge to new areas within financial econometrics.

    Skills - The candidate...

    • can analyze and deal critically with various sources of financial data;
    • understands the principles behind the construction of financial data series, such as trading days and returns;
    • is able to explain what makes financial data different from other types of data;
    • is able structure an empirical financial problem, and formulate a research strategy by selecting the relevant method to tackle the problem;
    • can understand econometric theory and apply this in the context of estimating the capital asset pricing model (CAPM), the Fama- French Model, and Event Study and other variations of the model;
    • can understand the limits of this method and has the ability to select more appropriate methods such as Panel Estimators, Time Series Techniques such as Unit Roots or Co-integration, or Instrumental Variable Techniques;
    • can understand the difference between a correlation and causal estimation strategy.

    Competence - The candidate...

    • can apply his/her knowledge and skills in new areas in order to carry out advanced assignments and projects;
    • can identify relevant ethical dilemmas in data collection and carry out his/her research with scholarly integrity.

  • Teaching

    Teaching

    We use a combination of lecture and case discussion every week. The lectures cover the econometric tools for each case; the class discussions reveal how to apply these techniques to a relevant problem. All students are expected to fully participate each week.

  • Recommended prerequisites

    Recommended prerequisites

    None. However, some students feel more comfortable to take the course after they have taken the mandatory courses in FIE. We also assume that students have taken a bachelor's course on basic econometrics at the level of MET4 (previously INT010) or comparable levels .

  • Credit reduction due to overlap

    Credit reduction due to overlap

    The course cannot be combined with BUS444 Økonometri for regnskap og økonomisk styring, BUS444E Econometrics for Business Research, BAN431 Econometrics and Statistical Programming, ECN402 Econometric Techniques, FIE401B Financial Data Analysis or FIE449 Financial Econometrics.

  • Requirements for course approval

    Requirements for course approval

    Mandatory Attendance: Attendance in all classes is mandatory for this course. Exemptions for not attending a specific class can be obtained from the lecturer on a case by case basis.

  • Assessment

    Assessment

    The final grade will consist of a 12 hours take home exam that is to be solved either individually or in a group of 2 students (70%), and oral participation during the course (30%). The exam and the course will be given in English , and the exam has to be answered in English.

    This exam can only be retaken under the same course code. You cannot retake the exam in FIE401B for this course.

  • Grading Scale

    Grading Scale

    Grading scale A - F.

  • Computer tools

    Computer tools

    Students will be expected to undertake an applied econometric analysis each week using a statistical software package.

  • Semester

    Semester

    Spring.

  • Literature

    Literature

    Cases must be purchased by the students from Harvard Business School, if they are not written by the lecturers. Also, there will sometimes be background reading set from introductory textbooks.

Overview

ECTS Credits
7.5
Teaching language
English.
Semester
Spring

Course responsible

Carsten Bienz, Department of Finance and Liam Brunt, Department of Economics (Spring Semester)