Structure, dynamics and predictability in techno-socio-economic systems


Loading...

Author / Producer

Date

2023-05

Publication Type

Habilitation Thesis

ETH Bibliography

yes

Citations

Altmetric

Data

Abstract

This habilitation thesis is focused on the analysis of modern techno-socio-economic systems, that are composed of a large number of non-linear interacting agents with adaptive behavior. Part I is focused on the stochastic spreading dynamics of SIR, SEIR type on network structure, and the interplay i.e. mapping between structure and dynamics. Chapter 1 provides details of a computational and analytical spreading framework, that can accurately and efficiently model (non)-Markovian stochastic interactions of a large number of agents in a network setting. Chapter 2 demonstrates that in the noisy setting of measurement testing and stochastic process, there are limits to the ability to infer the true number of infected nodes. Part II focuses on the adaptive financial complex systems with a timescale of minutes and hours, where we analyze aggregated data of intentions to buy or sell a specific amount of assets at a certain price of a large number of agents. Chapter 3 focuses on the intraday short-term volume $x_t$ forecasting in cryptocurrency markets. The predictions are built by using transaction and limit order book data from different markets where the exchange takes place. A temporal mixture ensemble model is proposed, capable of adaptively exploiting, for the forecasting, different sources of data and providing a volume point estimate, as well as its uncertainty. Chapter 4 focuses on intraday short-term volatility forecasting in the cryptocurrency market for price changes. In Chapter 5, we show how information from social media at a minute-level frequency can be used for estimating and forecasting the volatility of the Bitcoin/USD pair. Part III focuses on the analysis of predictability in noisy socio-technical systems. In Chapter 6, results suggest that easily interpretable methods like the Euler method, a model-free local-derivative-based forecasting benchmark, provide an effective alternative to more complex epidemic forecasting frameworks on short-term forecasting horizons. In Chapter 7, we study the predictability of bankruptcy, where besides financial data for each municipality, we use alternative institutional data along with the socio-demographic and economic context.

Publication status

published

External links

Editor

Book title

Journal / series

Volume

Pages / Article No.

Publisher

ETH Zurich

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Complex system; Complex networks; Dynamics; Social systems; ECONOMIC SYSTEMS; Socio-technical System

Organisational unit

03784 - Helbing, Dirk / Helbing, Dirk check_circle

Notes

Funding

Related publications and datasets