Finance and quantitative methods for economics
Study-unit Code
Statistical data science for finance and economics
Barbara Guardabascio
  • Barbara Guardabascio
  • 42 ore - Barbara Guardabascio
Course Regulation
Coorte 2022
Learning activities
Matematico, statistico, informatico
Academic discipline
Type of study-unit
Obbligatorio (Required)
Type of learning activities
Attività formativa monodisciplinare
Language of instruction
The quantitative estimation and evaluation of time series is an essential feature of the study and application of economics and finance. We will provide methodological background and data-analysis skills for time series analysis together an intuitive treatment of some elements of basic time series econometric in conjunction with the practical application of the methods to economic issues. More in details, we emphasize the measurement and interpretation of empirical results using time series data. A strong performance in this module is very important for students planning to take the econometrics or dissertation modules in final year.
The aim of the module is to familiarize you with the techniques economists use to measure economic relationships in order to test hypotheses, forecast, or assess the impact of economic policy. This is a very challenging field of applied economics. We do this in a practical way also using suitable computer software to investigate empirical relationships studied throughout the module.

Reference texts
Reading material on each course topic (handouts, slides, data sets, R scripts), will be made available to the students by the course instructors during the course.
Suggested books are:
Jeffrey M. Wooldridge (2020), “Introductory Econometrics: A modern approach”, 7th Edition, Cengage Learning.

George E. P. Box, Gwilym M. Jenkins, Gregory C. Reinsel and Greta M. Ljung, (2015) “Time Series Analysis: Forecasting and Contro”, 5th Edition,. John Wiley and Sons Inc.

Peter J. Brockwell and Richard A. Devis (2016) “Introduction to Time Series and Forecasting”, Springer.

Klaus Neusser (2016), “Time Series Econometrics”, Springer.
Educational objectives
By the end of the module, students should:
Have a critical understanding of the nature of economic and financial time series;
Be able to abstract the essential features of an econometric issue or problem;
Be able to perform and critically evaluate model adequacy using relevant diagnostic and specification tests;
Have knowledge of different types of dynamic empirical model and their important role in the analysis of time series econometrics.
The advanced statistical methods introduced in the course are strongly dependent on basic mathematical and statistical tools, therefore it is expected that students had successfully attended and completed the planned mathematics and statistics courses.
Teaching methods
The course is carried out through lectures and practicums. Techniques will be introduced by examples and described in mathematical formulas. Focus will be on the practical implementation of each technique, and interpretation of results.
Other information
Attending classes is strongly advised.
Learning verification modality
Assessment will be through a written examination, containing both theoretical questions numerical data analysis exercises.
The examination is designed to test your analytical skills and your interpretation and understanding of relevant issues that arise in time series work.
Extended program
1. Introduction to time series
Time series definition. OLS assumptions with time series variables. The concept of memory and persistence of time series. The concept of autocovariance and autocorrelation. Interpreting the confidence interval for autocorrelation. The “Q-statistics”
Stochastic process and its main characteristics
2. The stochastic process. Stationarity and stationary process. The White noise. Autocorrelation and autocorrelation function.
3. Autoregressive-moving average models ARMA (p,q)
The Lag Operator. The autoregressive models AR(p) – Mean, variance, autocovariance and autocorrelation.
Stationarity conditions. The moving average process MA(q) - mean, variance, autocovariance and autocorrelation. Condition of invertability. Autoregressive-moving average models ARMA (p,q).
4. Coefficient estimation in ARMA (p,q) processes.
Coefficients estimation in autoregressive models. Coefficient estimation in ARMA processes. Quality of adjustment of time series models. AIC information criterion. The concept of forecasting. Using AR model to forecast. Forecasting error and forecast intervals. The impulse response function.
5. Multivariate time series models
The multivariate time series. The VAR model (definition, stationarity, estimation and forecasting).
The concept of Granger causality.
6. Nonstationary models
Trend stationary models (definition, estimation and forecasting).
Difference stationary models. The Random Walk (mean, variance, autocovariance and autocorrelation). The Random Walk with drift. The Integrated process. The effect of a shock in a Random Walk series. Forecast with an Unit Root process.
7. The unit root problem
The unit root problem. Spurious trends and regressions. Unit root tests (Dickey-Fuller). ADF test and the choice of the number of lags.
8. Spurious Regression
The concept of spurious regression as a pitfall of time series regression. How to identify a spurious regression. The combination of stochastic processes. The Cointegration.
Condividi su