Econometric tests and date-stamping methods for financial bubbles

Abstract

In the modern economy, the role of financial markets is very important: thanks to them, the numerous financial assets that allow the global economic system to function are exchanged and allocated to the use deemed most efficient. However, it is well known that phenomena, known as speculative bubbles, often occur on the markets, which bring prices to values that do not correspond at all to the real value of the asset, generating problems not only for the efficient functioning of the markets themselves but also for the economy. This work addresses this issue at both a theoretical and an econometric and practical level. It is structured as follows: in the first chapter we will deal with the theoretical aspect of financial bubbles, focusing on the academic debate regarding their definition, origins, classification, and possible effects. Well-known historical examples of this phenomenon will be presented in the second chapter, to then move on to the econometric treatment of the topic: in the third chapter we will show some tests designed to detect speculative bubbles and to define a start and an end date. Finally, in the fourth chapter they will be applied to some time series to verify the results. In particular, we will consider three cases to date the phase of exuberance of 1929 and verify the possible presence of a bubble in the global stock market and especially in the technology sector, a widespread and much discussed idea. Furthermore, thanks to these analyses it will be possible to examine the differences between the results produced by the tests and procedures used.