Statistical models for economic time series analysis: decomposition methods; moving averages; exponential smoothing models; AR, MA, ARMA, ARIMA and seasonal ARIMA models. Consumer price indexes: theory and practice.
Lecture material is posted on the Moodle class web page.
A useful reference book is:
Di Fonzo T. e F. Lisi (2005), Serie storiche economiche. Analisi statistiche e applicazioni, Carocci Editore, Roma.
The list of the sections to be skipped is posted on the Moodle class web page.
Learning Objectives
The students are expected to master the basic concepts of time series analysis in the time domain and the basic concepts of consumer price indices.
In particular, the course aims to provide the following knowledge, competencies and skills.
Knowledge and understanding: theoretical foundations of the linear time series statistical models and the methodological bases of the Istat's consumer price indices and inflation measures. Official statistical sources for economic time series and consumer price indices and related metadata.
Application of knowledge and understanding: the student will acquire the methodological bases that make him able to apply the statistical procedures on time series through the use of statistical software and to interpret the results, dedicating particular attention to the nature and reliability of the analyzed data and to the potentialities and limits of the methods used. Through the drafting of the report (see the section “Type of assessment”) he will demonstrate its communication skills in a technical/scientific context.
He will be able to interpret in a correct way the inflation indicators diffused by Istat.
Prerequisites
It is assumed that students are familiar with basic descriptive and inferential statistics (topics covered in B018993-STATISTICA).
Teaching Methods
In addition to classroom lectures, the course includes computer labs.
Further information
In order to get access to the Moodle class web page students are required to send an e-mail request to the teacher.
Type of Assessment
The exam consists of an oral interview.
Before the examination it is necessary to prepare a written report containing the statistical analysis of an economic time series. The instructions on how to write the report and the timing to be fulfilled for delivery can be found on the Moodle web page of the course.
During the oral examination the report will be discussed and questions will be asked about the whole program, with at least one question on each of the points 2, 3, 4, 5, 6 and 8 specified in the next section.
Course program
1. Time series analysis in economics.
Introductory univariate time series analysis with linear methods.
2. Exploratory analysis: plots, summary statistics, transformations (logs, differencing, index numbers), sample autocorrelation.
3. Time series decomposition. Time series components (trend, cycle, seasonal component and error).
4. Moving averages. Census I seasonal decomposition.
5. Exponential smoothing. General introduction. Simple exponential smoothing. Holt’s linear trend method. Holt-Winters’ seasonal method.
6. Stochastic processes. Wold’s theorem. AR, MA, ARMA, ARIMA and SARMA models. Box-Jenkins methodology.
7. Computer labs with applications on real time series.
8. Consumer price indexes. Istat indexes (NIC, FOI; IPCA). Price collection and calculation method. Measures of inflation. Core inflation. Perceived inflation.