Kleanthis Natsiopoulos

Kleanthis Natsiopoulos
University of Thessaly | UTH · Τμήμα Οικονομικών Επιστημών

Ph.D. in Economics


How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
I am a Ph.D. Candidate in Economics, specialized in Econometrics and Computational Statistics, and a Machine Learning enthusiast. In the past, I have worked as a Data Analyst in the Oil and Shipping industry. My skills and interests include statistical modeling, machine learning, econometrics, time-series analysis, and computational economics.
September 2016 - December 2017
Athens University of Economics and Business
Field of study
  • Statistics
September 2011 - October 2015
University of Thessaly
Field of study
  • Economic Science


Publications (3)
This paper replicates the UK earnings equation using the autoregressive distributed lag (ARDL) modeling approach and the bounds test for cointegration by Pesaran et al. (Journal of Applied Econometrics, 2001, 16(3), 289–326). The findings from the narrow sense fully replicate the original results using the open source language R and the ARDL packag...
The purpose is to explore the cointegrating relationships between the 1% top income share and the macroeconomic factors of credit, education, GDP, inflation, population growth and trade in order to reveal if there is a long-run relationship. This relationship is tested for four different countries: Greece, France, USA and UK. We are trying to see i...
Conference Paper
Full-text available
An indirect but at the same time efficient solution to the problem of environmental degradation in terms of biodiversity loss is the one supported by the theory of the Environmental Kuznets Curve (EKC). As the theory indicates, the economic growth of a country through sustainable development is able to support the biodiversity conservation and redu...


Question (1)
I need to construct a poisson regression but the problem is that the exposure variable contains zeros and I can't take the log.
For example, my dependent variable is the number of successes, and the exposure variable the number of events.
This way, where the number of successes are 0 the corresponded number of events are also zero, and it's not possible to take the log(offset) as exposure variable.
How could I possibly manage to model this?
Any R code on this?
Thank you in advance.