Back to search

Institutional Repository of Vilnius University

Procesai, apibūdinami q-Eksponentiniais laukimo laikais /

Abstract

dc:description

This bachelor thesis presents the analysis of a process characterized by q-exponential waiting times. Empirical data of the limit order book of 9 NASDAQ-traded stocks AMZN, NVDA, HD, NFLX, MA, LLY, TSLA, ADBE and JNJ, which are obtained from the Limit Order Book (LOB) data system LOBSTER, are selected as the research object. The purpose of this work is to study LOB data and determine the statistical properties of order lifetimes, i.e. distribution. Since the empirical data of the financial market contains deviations from their exponential nature, in this paper the lifetimes of orders are considered to be the product of a non-extensive complex system, therefore, the basics of Tsallis statistics are applied in the analysis. In different 2020 August trading day periods and for the 9 different stocks the calculated distributions of order lifetimes are marginally different, suggesting that the order lifetimes are a sample of the population that follows the theoretical distribution. By applying the Maximum Likelihood Estimation (MLE) method, the parameters of the q-exponential distribution of order lifetimes were calculated, with the averages of 9 stocks being λ_q=0.24, q=1.49. To determine the suitability of the q-exponential distribution for describing the statistical properties of order life series, the data were compared with the Weibull distribution. After calculating the Mean Square Error (MSE) and performing the Kolmogorov-Smirnov test, it is shown that the time series of order lifetimes belong to a population that is q-exponentially distributed, because out of the two studied distributions, the p-value of 0.05 was exceeded only by the q-exponential distribution with an average p-value of 0.53 of 9 stocks. The established statistical properties of empirical data order lifetimes can be used for further modeling of transaction flow imbalance in complex social systems. The created models will help to understand the second-order statistical properties of the data and the origin of the autocorrelation of the time series. The title of this bachelor thesis is "Processes characterized by q-exponential waiting times". This thesis is prepared by Vilnius University student Gediminas Jonikas and supervised by manager Doc. Habil. Proc. Vygintas Gontis – leader of complex physical and social systems group.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Jonikas, Gediminas,

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
lit

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Jonikas, Gediminas,. Procesai, apibūdinami q-Eksponentiniais laukimo laikais /. Institutional Repository of Vilnius University, 2023. https://repository.vu.lt/VU:ELABAETD192833443&prefLang=en_US