journal.pone.0270492.pdf (3.56 MB)
An empirical data analysis of "price runs" in daily financial indices: dynamically assessing market geometric distributional behavior
journal contribution
posted on 2023-06-10, 04:05 authored by Hector Raul Olivares-Sanchez, Carlos Manuel Rodriguez-Martinez, Hector Francisco Coronel-Brizio, Enrico Scalas, Thomas Henry Seligman, Alejandro Raul Hernandez-MontoyaIn financial time series there are time periods in which market indices values or assets prices increase or decrease monotonically. We call those events "price runs", "elementary uninterrupted trends" or just "uninterrupted trends". In this paper we study the distribution of the duration of uninterrupted trends for the daily indices DJIA, NASDAQ, IPC and Nikkei 225 during the period of time from 10/30/1978 to 08/07/2020 and we compare the simple geometric statistical model with p = 1/2 consistent with the EMH to the empirical data. By a fitting procedure, it is found that the geometric distribution with parameter p = 1/2 provides a good model for uninterrupted trends of short and medium duration for the more mature markets; however, longest duration events still need to be statistically characterized. Estimated values of the parameter p were also obtained and confirmed by calculating the mean value of p fluctuations from empirical data. Additionally, the observed trend duration distributions for the different studied markets are compared over time by means of the Anderson-Darling (AD) test, to the expected geometric distribution with parameter p = 1/2 and to a geometric distribution with a free parameter p, making possible to assess and compare different market geometric behavior for different dates as well as to measure the fraction of time runs duration from studied markets are consistent with the geometric distribution with p = 1/2 and in parametric free way.
History
Publication status
- Published
File Version
- Published version
Journal
PLoS OneISSN
1932-6203Publisher
Public Library of ScienceExternal DOI
Issue
7Volume
17Article number
e0270492Department affiliated with
- Mathematics Publications
Research groups affiliated with
- Probability and Statistics Research Group Publications
Full text available
- Yes
Peer reviewed?
- Yes
Legacy Posted Date
2022-06-27First Open Access (FOA) Date
2022-06-27First Compliant Deposit (FCD) Date
2022-06-27Usage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC