PRICE JUMP DYNAMICS

Price jump dynamics and evolution of market panic

 Partecipanti

Mappa

 Word cloud

Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.

 Obiettivo del progetto (Objective)

'The proposed project studies the dynamics of extreme price movements, or price jumps, using high-frequency data of financial assets. In the first stage of the project, fellow address the issues of price jump indicators in the multivariate context as opposed the univariate price jump indicators dominating the literature. Then, he will compare his indicators with those in the literature using Monte Carlo analysis as well as real data. In the second stage, he focuses on the price jump dynamics methodology, which is rarely touched in the literature. First, he develops the two-stage method, where he first identifies price jumps using the previous mentioned indicators, and, in the second, step he estimate the market dynamics in different market regimes defined through the presence of jumps. He further extends the model and joins both stages into one. This new model based on the maximum likelihood will estimate price jumps and their dynamics in one step and thus provide more robust results. He further applies the methodology on the simulated data and the empirical high-frequency data for various stocks or stock market indices. Thus, he obtains the quantitative description of phenomena like information spread across financial markets—price jumps serves as a proxy for moment when information hits the market—or even market panic. The results can be directly applied to study financial integration of different markets, their stability in the financial crisis, contagion of market panic and to propose more efficient policies and construct more robust portfolios. Finally, fellow proposed a theoretical model based on the continuous-time DSGE methodology, which set the theoretical grounds for the observed phenomena. When fitted on the data, the model will significantly extend the current understanding of the financial markets and provides new theoretical methodology.'

Introduzione (Teaser)

In an attempt to manage financial risk and the volatility of assets within certain financial markets, studies are being conducted to further understand price jump dynamics. Through the study of high-frequency financial time series, more comprehensive and accurate price jump models have been developed.

Descrizione progetto (Article)

In the wake of the recent financial crisis, understanding phenomena such as market panic and predicting market alteration has become even more important. Price jumps evidently influence financial markets, most practically in terms of investment gains and losses. Improving our understanding of price jumps, and the factors that influence them, can help predict future market behaviour and allow for more effective risk management.

The EU-funded project 'Price jump dynamics and evolution of market panic' (PRICE JUMP DYNAMICS) focused on analysing price jumps, using high-frequency data of financial assets. There were three main results.

First, co-jumps and co-arrivals were introduced within the co-features framework, and the proposed framework illustrated through high-frequency data. Second, a methodology was created to identify commonalities defined in terms of co-arrivals and co-jumps. While determined at high frequency, the commonalities were looked at as results of low-frequency macro-factors or predictors. Third, the project proposed a framework for predicting European price jump arrivals and identifying the significant factors.

New empirical evidence was discovered, showing that emerging markets in Central and Eastern Europe have a delayed reaction to news announcements. It was also discovered that foreign macroeconomic news mainly accounts for price jumps in these areas, largely influenced by markets in the United States.

Researchers also conducted a focused analysis of the Prague Stock Exchange (PSE) and the New York Stock Exchange (NYSE). This revealed the PSE doesn't react long-term to financial distress or credit default swap movements, while the NYSE reaction to both is sector/company-specific.

The project also argued that the relationship between price jumps, Gaussian variance and financial transaction taxes (FTTs) is crucial to understanding the frequency of catastrophic market events. The agent-based model results showed that FTTs may increase the variance while decreasing the impact of price jumps. Analysis of foreign exchange markets suggested price jumps can serve as a tool for identifying temporal market inefficiencies that arise due to extensive hedging around news announcements.

PRICE JUMP DYNAMICS has provided significant advancements in the analysis of price jumps within the high-frequency financial time series. Such knowledge will aid various areas of the financial sector, from regulatory measures to the trading floors. Hopefully, this knowledge will provide protection against market volatility and improve risk management, preventing the catastrophic consequences of market panic.