Gifuni, Luigi (2022) Essays in Empirical Macroeconomics: Frequentist and Bayesian Analysis. [Tesi di dottorato]

Preview

Text Gifuni_Luigi_33.pdf Download (3MB) | Preview

[error in script] [error in script]

Item Type:	Tesi di dottorato
Resource language:	English
Title:	Essays in Empirical Macroeconomics: Frequentist and Bayesian Analysis
Creators:	Creators Email Gifuni, Luigi Luigi.Gifuni@unisa.it
Date:	21 January 2022
Number of Pages:	100
Institution:	Università degli Studi di Napoli Federico II
Department:	Scienze Economiche e Statistiche
Dottorato:	Economia
Ciclo di dottorato:	33
Coordinatore del Corso di dottorato:	nome email Pagano, Marco pagano56@gmail.com
Tutor:	nome email Simonelli, Saverio UNSPECIFIED
Date:	21 January 2022
Number of Pages:	100
Keywords:	Securities markets programme, outright monetary transactions, longer-term refinancing operations, event-study, multi-country vector autoregression, Gibbs sampling, Brexit, global vector autoregression, bootstrap after bootstrap, Metropolis-Hastings algorithm.
Settori scientifico-disciplinari del MIUR:	Area 01 - Scienze matematiche e informatiche > MAT/02 - Algebra Area 01 - Scienze matematiche e informatiche > MAT/06 - Probabilità e statistica matematica Area 13 - Scienze economiche e statistiche > SECS-P/02 - Politica economica Area 13 - Scienze economiche e statistiche > SECS-P/05 - Econometria
Date Deposited:	20 Apr 2022 05:43
Last Modified:	07 Jun 2023 11:10
URI:	http://www.fedoa.unina.it/id/eprint/13530

Collection description

This work presents two empirical studies aimed to evaluate the financial and macroeconomic effects following monetary policy and political decisions. Both studies rely on the application of an event-study methodology, which has long been considered an accurate statistical procedure to quantify the impact effects following a broad range of events, such as political decisions, implementation of new business strategies, exogenous geopolitical events, and many others. The outline of this thesis is structured as follows. Chapter 1 expands the empirical work of Altavilla et al. (2014) by considering the impact of alternative non-conventional monetary policies (Securities Markets Programme, Longer-Term Refinancing Operations and Outright Monetary Transactions) on multiple financial assets: (i) benchmark stock market prices, (ii) bank index prices, (iii) 10-year government yields. After assessing the economic impact of the aforementioned policies, we perform a multi-country vector autoregression in order to analyse the dynamic effects that a positive monetary shock generates on the gross domestic product and the harmonised index consumer price of 4 European countries (Germany, France, Italy and Spain). In contrast with Altavilla et al. (2014), and according with our empirical evidence, we conclude by inferring that not all non-conventional policies behave as exogenous positive monetary shocks, as Longer-Term Refinancing Operations has generated negligible financial effects on the economies considered. Chapter 2 investigates the economic impact effect on benchmark stock market prices and 10-year government yields of the main trading partners of the UK, after nine important Brexit events listed in Table 1. We cover a sample period running from the referendum result of June 24, 2016 through the general elections of December 12, 2019. The surprising Brexit outcome is considered as an exogenous uncertainty shock, and thus we investigate the effects that a UK uncertainty shock generates on the real economy of 15 alternative countries. Then, in the spirit of Brodersen et al. (2015), we also investigate the average macroeconomic effect of this political decision, by comparing two different scenarios: the case of Brexit and no-Brexit. The latter is represented by estimating a stochastic system from 1996Q1 through 2019Q1, and then forecasting the gross domestic product and the inflation rate for the next 3 years. In both studies when the number of predictors is objectively low, we use a frequentist econometric approach. Whereas, when the equations become richly parameterized, we apply econometric strategies based on Bayesian theory (Minnesota prior, Gibbs sampling, Metropolis-Hastings) in order to deal with the curse of dimensionality.

Downloads

Actions (login required)

View Item