• Detect situations in which the ordinary least squares estimator is not adequate and be able to explain why
• Choose an econometric model from those introduced in the course and explain why it is the suitable model for the specific situation
• Interpret estimation results in STATA / R output correctly and comment on appropriateness of their presentation
• Relate STATA / R code and STATA / R output to the econometric models introduced in the course
• Work with STATA / R to do econometric analysis with the models introduced in the course

With advances in computer technology, electronic systems have replaced paper files to a large extent in most firms. A wealth of data is being produced in operations by all sorts of business units. This includes internal pay roll and personal records in Human Resources and Finance, production data on the amount of in- and output of machines and customer data, including prices and quantity sold. All these data have in common that they are typically large enough in terms of number of observations and number of variables to conduct multi dimensional statistical analysis. While these data are increasingly available as a side product of operations, their use in firms is often limited because of a lack of quantitative data analytics skills. As a matter of fact the demand of leading international firms for graduates with profound skills in quantitative methods normally exceeds the supply. The goal of this course is to provide students with skills to analyse these data by means of modern econometric techniques. Thus, it will enable them to use these data in practise in order to move towards evidence based decision making within firms. An integral component of the course is therefore hands-on practical work with statistical software and effective presentation of statistical results to a general audience without background in quantitative methods. However, the main goal of the course is to provide a deeper understanding of the underlying econometrics using important subject content topics as motivating examples. These may include:

• Human Resources Management:

• Estimation of the determinants of salaries. Are the salaries of top executives mainly related with firm characteristics?

• It is often observed that females earn less on average than males. It can have important legal consequences for a firm if they pay females systematically lower for the same work only because they are female. We present various ways to estimate whether there is gender discrimination among staff.

• Larger firms offer various optional benefits such as holiday buy schemes or fitness vouchers to their employees. It is important for planning purposes to understand the determinants for the decision to choose these benefits.

•  Marketing:

• Estimating the effect of a marketing initiative on product demand using sales and marketing data.

• Estimating the determinants for the decision to purchase a product.

•  Production:

• Estimation of production or cost functions using input and output data.

• Estimating the effect of staff training on production output.

 The course will first deepen the understanding of estimation of the linear regression model by Ordinary Least Squares. It will then cover methods which can be used if the key variables of interest are not exogenous, such as instrumental variable techniques or panel data techniques. This is followed by extending the linear model to a model with a limited dependent variable, which is estimated by Maximum Likelihood. It may also cover topics such as decomposition techniques or estimation of systems of equations. The course will provide code for the statistical softwares STATA  and R. Students can choose between R and Stata.

By doing the course a student will

• practice the work with multivariate data;
• broaden their repertoire of statistical methods relevant for the analysis of firm data;
• develop the ability to work practically with  statistical models;
• become proficient in the use of STATA / R statistical software;
• obtain the required methods skills to do multiple regression analysis in an empirical Master dissertation.

Lectures:

    Lecture notes
    Selected scientific articles to be specified during the course
    Further recommended readings, revision material and articles will be posted on Canvas

Textbook:

Wooldridge, J. (2020), "Introductory Econometrics", 7th edition, Cengage, ISBN10: 1-337-55886-9