Polimi Fintech Series – Michele Azzone – January 18, 2021

Jan 18 2021
The Polimi Fintech Series, under the fintech-ho2020.eu and the Cost Fin-AI.eu project, presents
 
January 18th, 2021 – 17.30 (CET)
 
Virtual room: Click here to access the Zoom Virtual Room, or insert the Meeting id on your Zoom app: 824 7266 9724
 
M. Azzone (Politecnico di Milano)
with E. Barucci, G. Giuffra and D. Marazzina
 
A Machine Learning Model for Lapse Prediction in Life Insurance Contracts
Abstract:
In this work, we use the Random Forest methodology to predict the lapse decision of life contracts by policyholders. The methodology outperforms the classical logistic model in describing the phenomenon. We use global and local interpretability tools to investigate how the model works. We show that non economic features (time passed from the incipit of the contract and the time to expiry, as well as the insurance company) play a significant effect in determining the lapse decision while economic/financial features (except the disposable income growth rate) play a limited effect. The analysis shows that linear models, such as the logistic model, may not be adequate to capture the heterogeneity of financial decisions.
 
Stay updated with latest news on QFinLab seminars subscribing here: https://bit.ly/2TNtC6e
 
Information about future seminars: www.qfinlab.polimi.it/all-news

Al via la nuova edizione edizione di Policollege

QFinLab partecipa per la terza volta al Policollege con il corso Primi Passi nella Finanza Matematica.
 
PoliCollege è un progetto di didattica innovativa che offre agli studenti delle scuole secondarie di secondo grado di tutta Italia l’opportunità di approfondire e ampliare le proprie conoscenze tecnico-scientifiche seguendo corsi online tenuti da docenti del Politecnico di Milano
 
 
Il corso Primi Passi nella Finanza Matematica, erogato dal gruppo QFinLab, ha l’obbiettivo di fornire agli studenti gli strumenti per rispondere a domande concrete: cos’è lo spread? Come viene calcolato? Quali sono gli elementi da prendere in considerazione per chiedere un prestito? E ancora: quali sono le condizioni cui porre attenzione nell’aprire un conto corrente? Quali gli investimenti più vantaggiosi? Quanto è oneroso acquistare un oggetto a rate?
 
La terza edizione partirà l’11 gennaio.
 
Per maggiori informazioni su Policollege: https://www.policollege.polimi.it/

Summer School – From Networks to Neural Networks in Finance

Jun 14 2021

From Networks to Neural Networks in Finance

Lake Como School of Advanced Studies – 14-18 June 2021

The School aims to present the state of the art on methodologies and applications of Neural Networks and Nets to finance. The expected audience of the school is provided by PhD student and young researchers interested in applications of Neural Networks and Nets to finance.

Program

The school is organized through three different initiatives:

  • Four minicourses
  • Lectures
  • Workshops with students and participants to the school on developing research ideas.

Minicourses:

  • Albert Diaz Guilera, Universitat de Barcelona (12 hours):
  • Tomaso Aste UCL, London (6 hours): Information filtering networks for socio-economic systems
  • Matteo Matteucci, Politecnico di Mialno (12 hours): Introduction to neural networks: from theory to practice
  • Josef Teichmann, ETH Zurich (6 hours) (TBC)

Lectures:

  • Christoffer Kok, European Central Bank, Contagion modelling at the ECB: analytical frameworks and policy usage
  • Paolo Giudici, Università di Pavia, Network based credit risk models for peer to peer lending.
  • Andrea Prampolini, Intesa Sanpaolo, Limit order book simulation with interactive agents

 

Detailed program: https://nnnf.lakecomoschool.org/program/

Speakers: https://nnnf.lakecomoschool.org/speakers/

Application: https://nnnf.lakecomoschool.org/application/

 

School Director:

Emilio Barucci
Dipartimento di Matematica
Politecnico di Milano
emilio.barucci@polimi.it

Scientific Committe:

Emilio Barucci, Roberto Baviera, Daniele Marazzina.

Organizing Committee:

Michele Azzone, Emilio Barucci, Roberto Baviera, Matteo Brachetta, Giancarlo Giuffra, Francesca Grassetti, Daniele Marazzina.

 

Website: https://nnnf.lakecomoschool.org/


Polimi Fintech Series – Jeremy D. Turiel – December 09, 2020

Dec 09 2020
The Polimi Fintech Series, under the fintech-ho2020.eu and the Cost Fin-AI.eu project, presents
 
December 9th, 2020 – 17.30 (CET)
 

Virtual room: Click here to access the Zoom Virtual Room, or insert the Meeting id on your Zoom app: 826 4084 3401

 
J. D. Turiel (UCL-ICL, Barclays Investment Bank)
with A. Briola and T. Aste
 
DEEP LEARNING MODELLING OF THE LIMIT ORDER BOOK: A COMPARATIVE PERSPECTIVE
 Abstract:
We address theoretical and practical questions in the domain of Deep Learning for High Frequency Trading. State-of-the-art models such as Random models, Logistic Regressions, LSTMs, LSTMs equipped with an Attention mask, CNN-LSTMs and MLPs are reviewed and compared on the same tasks, feature space, and dataset and clustered according to pairwise similarity and performance metrics. The underlying dimensions of the modelling techniques are hence investigated to understand whether these are intrinsic to the Limit Order Book’s dynamics. We observe that the Multilayer Perceptron performs comparably to or better than state-of-the-art CNN-LSTM architectures indicating that dynamic spatial and temporal dimensions are a good approximation of the LOB’s dynamics, but not necessarily the true underlying dimensions.
 
Click here to download the paper
 
Stay updated with latest news on QFinLab seminars subscribing here: https://bit.ly/2TNtC6e
 
Information about future seminars: www.qfinlab.polimi.it/all-news

Polimi Fintech Series – Emilio Barucci – November 09, 2020

Nov 09 2020
The Polimi Fintech Series, under the fintech-ho2020.eu and the Cost Fin-AI.eu project, presents
 
November 9th, 2020 – 17.30 (CET)
 

Virtual room: Click here to access the Zoom Virtual Room, or insert the Meeting id on your Zoom app: 872 7241 8663 

 

E. Barucci (with M. Bonollo, F. Poli, E. Rroji)
A machine learning algorithm for stock picking built on information based outliers
 Abstract:
We build an algorithm for stock selection based on indicators of time series of stocks (return, volume, volatility, bid-ask spread) that should be associated with the dissemination of private information in financial markets. We use a machine learning algorithm for the identification of the most relevant indicators for the prediction of stock returns and to define a trading strategy. The procedure combines a sequential inclusion of predictors with a classification algorithm for the trading signal. We apply the methodology to two sets of stocks belonging respectively to the EUROSTOXX50 and the DOW JONES index. Performance is smoother than in the Buy&Hold strategy and yields a higher risk-adjusted return, in particular in a turbulent period. However, outperformance vanishes when 5-10% transaction costs are inserted.
 
Click here to download the paper
 
Stay updated with latest news on QFinLab seminars subscribing here: https://bit.ly/2TNtC6e
 
Information about future seminars: www.qfinlab.polimi.it/all-news