GROUP READING OF FACULTY OF INTERNATIONAL BUSINESS
10/01/2023
Group reading session of the Faculty of International Business took place with the topic "Time series regression and applications in international business research" presented by Huynh Thi Dieu Linh, PhD.
At the beginning, Huynh Thi Dieu Linh, PhD introduced the regression model with time series data. Time series data are defined as ordered observations in time, which cannot be reordered at random. In regression analysis with time series data, a very important assumption is that the time series under consideration is stationary. If a sequence is non-stationary, we can study its behavior only for the period under consideration. Besides, if we have two or more nonstationary series, regression analysis with such series can lead to spurious regression or nonsense regression.
According to Huynh Thi Dieu Linh, PhD, a time series is stationary if its mean and variance are constant over time and the covariance between two periods depends only on the distance between the two periods. period rather than the actual time at which the covariance is calculated. Specifically, there are three basic ways to investigate the stationarity of a time series: (1) graphical analysis, (2) autocorrelation plots, and (3) unit root analysis.
The reading group activity has opened up a lot of attention and discussions for the participants.