Dr Danilo Petti
-
Email
d.petti@essex.ac.uk -
Telephone
+44 (0) 1206 876292
-
Location
2.410, Colchester Campus
Profile
Biography
Danilo Petti undertook his undergraduate study in Economics at the University of Salerno to then continue with an MSc in Statistics and Actuarial Science. He graduated with an MSc in Statistics with Distinction at University College London to then complete a PhD in Statistics at the University of Salerno. During his doctoral years, he simultaneously followed and successfully completed the course in Advanced Statistics and Actuarial Science at the University of Florence. He is also a certified trainer (score 980/1000) of Statistical Analysis System (SAS). He is now interested in various research areas in statistical methodology and applications, including computational statistics, statistical learning methods, graphical models, causal inference, survival analysis and copula models.
Qualifications
-
BSc In Economics ( 110 cum Laude) University Of Salerno,
-
MSc in Statistics and Actuarial Science ( 110 cum Laude) University of Salerno,
-
MSc in Statistics (Distinction) University College London,
-
PhD in Statistics University of Salerno,
Appointments
University of Essex
-
Lecturer in Statistics and Data Science, Mathematical Science, University of Essex (16/1/2023 - present)
Other academic
-
SAS Trainer, Department of Economics and Statistical Science, University of Salerno (1/10/2020 - 15/12/2022)
Research and professional activities
Research interests
Copula link-based additive models for bivariate time-to-event outcomes
In medical research, the relationship and timing between two events, such as multi-organ diseases or maculopathy, are often of interest. However, it's not always possible to pinpoint exactly when these events occur due to missing information, a situation known as censoring. Censoring commonly occurs in case-control studies that extend over several years. Our objective was to develop a versatile tool capable of analyzing these event pairs despite the gaps in data. This tool had to be flexible enough to accommodate various types of missing information, for instance, when a patient discontinues their participation midway through a study, joins after the study has begun and leaves before its completion, or reaches the end of the study without the disease manifesting.
Variable Ranking in Bivariate Domain
In studies where we observe how long it takes for certain events to happen, it's essential to identify which factors are influential. However, this becomes increasingly difficult as the amount of data grows. Our goal was to develop a new method that can pinpoint these important factors effectively, even when dealing with a large amount of information and when some data might be missing or incomplete.
Analysis of Financial Time Series
The global financial ecosystem recently faced a significant challenge with the collapse of major institutions like Silicon Valley Bank, leading to widespread fear among investors. This event highlighted the potential for a systemic crisis, characterized by liquidity constraints and heightened market volatility. Amidst this turmoil, Bitcoin has emerged as a novel asset class, offering investors an alternative means of diversification. In response to these developments, our goal was to understand the impact of the banking crisis on Bitcoin and its potential as a stabilizing force in the face of global financial uncertainty. To achieve this, we set out to develop a banking index that could help isolate the effects of the banking crisis on Bitcoin and then analyze the resulting market volatility.
Development of Statistical Softwares in R Environment
Development of statistical software in the R environment.
Conferences and presentations
BRBVS: Bivariate Variable Ranking In copula survival Model(s)
COMPSTAT 2024, 26th International Conference on COMPUTATIONAL STATISTICS (COMPSTAT 2024), Giessen, Germany, 28/8/2024
Variable ranking in bivariate copula survival models
Invited presentation, CLADAG 2024, Fisciano, Fisciano, Italy, 11/9/2023
Teaching and supervision
Current teaching responsibilities
-
Linear Regression Analysis (MA317)
-
Applied Statistics (MA321)
-
Capstone Project: Mathematics (MA829)
-
Capstone Project: Mathematics (MA830)
-
Capstone Project: Mathematics (MA831)
-
Advanced Capstone Project: Actuarial Science, Data Science or Mathematics (MA930)
Publications
Journal articles (2)
Petti, D. and Sergio, I., (2024). Bank Crisis Boosts Bitcoin Price. Journal of Risk and Financial Management. 17 (4), 134-134
Petti, D., Eletti, A., Marra, G. and Radice, R., (2022). Copula link-based additive models for bivariate time-to-event outcomes with general censoring scheme. Computational Statistics & Data Analysis. 175, 107550-107550
Conferences (1)
Petti, D., Niglio, M. and Restaino, M., (2023). Variable Ranking In Bivariate Copula Survival Models
Thesis dissertation (1)
Petti, D., (2023). Copula link-based additive models for bivariate time-to-event Outcomes with general censoring scheme: Computational advances and variable ranking procedures
Other (1)
Petti, D., Niglio, M. and Restaino, M., (2024).R Package BRBVS: Variable Ranking in Copula Survival Models Affected by General Censoring Scheme,CRAN
