Polimi Fintech Series

Nov 09 2020

QFinLab, the research group on Quantitative Finance – Politecnico di Milano, is pleased to announce Polimi Fintech Series.

Featuring contributions from both leading academics and practitioners, the series, under the fintech-ho2020.eu and the Cost Fin-AI.eu project, will explore challenges facing Fintech today. Guided by the expertise of QFinLab, this seminar series will provide a forum for discussion over technology applied to financial industry.

To stay updated with latest news on QFinLab seminars subscribe here: https://bit.ly/2TNtC6e

The first appointment with Polimi Fintech Series, on November 9th 2020 at 17.30 Italian time, will host Emilio Barucci (Politecnico di Milano) presenting

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.

More details about the first seminar will be announced soon.

Events are open to any interested parties. Due to Covid-19 emergency, seminars will be delivered online.

Stay updated with latest news on QFinLab seminars subscribing here: https://bit.ly/2TNtC6e

Conference Big Data and Machine Learning in Finance

Jun 10 2021

www.mate.polimi.it/fintech
June 10-11, 2021 – Online Conference

Big Data and Machine Learning are driving a significant transformation in the financial industry. Amazing examples include: robo-advisory; predicting frauds in payment systems; development of sophisticated algorithmic trading strategies; systemic risk assessment; rating of companies/financial products using a huge amount of information; development of chatbots for customers; nowcasting of financial time series; digital marketing; instant pricing of insurance products.

The transformation concerns the academia and the financial industry. The goal of the conference is to bring together academicians with different backgrounds (economists, finance experts, data scientists, econometricians) and representatives of the financial industry (banks, asset management, insurance companies) working in this field.

Papers on all areas dealing with Machine Learning and Big Data in finance (including Natural Language Processing and Artificial Intelligence techniques) are welcomed. The conference targets papers with different angles (methodological and applications to finance).

Invited speakers:

Tomaso Aste (University College London)
Emanuele Borgonovo (Università Bocconi)
Tucker Balch (JP Morgan AI research)
Juri Marcucci (Bank of Italy)
Georgios Sermpinis (Adam Smith Business School, University of Glasgow)

For information: www.mate.polimi.it/fintech

Seminar Niklas Wagner – May 12, 2020

May 12 2020

Niklas Wagner (Passau University)

Give Me a Break: Is the Equity Premium a Trading Break Premium?

May 12, 2020 – 12.30

Abstract

This paper addresses the relation between market risk and expected market returns under periodic trading breaks. We propose a model where asset prices are driven by a diffusion process that operates during the trading day and a separate process that captures overnight price changes. Our empirical analysis shows that both components are important in explaining the equity market risk premium.
Trading breaks entail a lack of market functionality and liquidity and our results reveal that investors ask for a premium to hold the market portfolio overnight. Including additional state variables into the model, we find that uncertainty risk and illiquidity risk are both priced as well.

Seminar Pasquale Cirillo – May 5, 2020

May 05 2020

Pasquale Cirillo (TU Delft)

The distortions of finance

May 5, 2020 – 12.30

Abstract

Finance is a world of distortions. Many tools we use, many findings we know are actually the result of a distortion.

Take the well-known Black-Scholes model: the probability to be in the money at maturity under P and Q is a distortion. And the price of a European call? Another distortion.

Consider risk management, think about the expected shortfall, and—guess what?—a distortion.

And if you think that copulas are immune, you are wrong, plenty of distortions there.

Model risk is often represented in terms of distortions.

So, let’s talk about distortions, and in particular about the special class of Lorenz transforms.

Canceled – Seminar Rosario Mantegna – March 24, 2020

Mar 24 2020

Sospeso per indicazioni ministeriali

Rosario Mantegna (Palermo University)

Trading networks in a stock exchange: the case of LSE and of the Nordic Stock Exchange with a focus on high-frequency trading

Canceled

Abstract

We study the heterogeneity of financial actors trading in a fully electronic stock market. The investigated stock markets are the Nordic Stock exchange and the London Stock Exchange. By using network concepts and network filtering methods we detect specialization of market members. Specifically, we detect over-expressed and under-expressed trading interactions of some market members that are statistically robust and that are stable over long period of times. These networked structures are specific to each market, time period, and specific financial asset. By analyzing trading networks both for buyers-sellers and aggressors-counterparties we interpret some network structures as related to the diffusion and implementation of high-frequency trading.