Tony Guida

Université Savoie Mont Blanc | UdS · Economy-Finance Area

In this article, we investigate the impact of truncating training data when fitting regression trees. We argue that training times can be curtailed by reducing the training sample without any loss in out-of-sample accuracy as long as the prediction model has been trained on the tails of the dependent variable, that is, when ‘average’ observations h...

Evolutionary algorithms are not new and have been developed, both concept and framework, around the 1950's on the idea that the evolution process could be used as general-purpose optimization tool. The goal of this paper is to propose an alternative to classical optimization techniques that can handle systems of a very high dimension. With the rapi...

In this article, we investigate the impact of truncating training data when fitting regression trees. We argue that training times can be curtailed by reducing the training sample without any loss in out-of-sample accuracy as long as the prediction model has been trained on the tails of the dependent variable, that is, when 'average' observations h...

This chapter proposes to benefit from the advantages of machine learning (ML) in general and boosted trees in particular, e.g. non‐linearity, regularization and good generalization results, scaling up well with lots of data. It gives a mildly technical introduction to boosted trees. The chapter introduces the construction of the dataset with the fe...

Machine learning continues to advance in more sophisticated ways in order to analyse aspects of life big data can record, from our physical well-being to our spending habits and driving behaviour. However, financial markets are a different kind of animal altogether, and while they may seem data rich, their complexity is extremely hard to be underst...

Recent criticisms suggest that Machine Learning‐based approaches only suite predicting very short‐term price movements. Tony Guida and Guillaume Coqueret apply well‐known ML algorithms to systematic equity investment, presenting a methodology which shows a critical stage of feature and label engineering, a stet that helps uncover hidden structures...

In this chapter, we apply a popular Machine Learning approach (extreme gradient boosted trees) to build enhanced diversified equity portfolios. A simple naïve equally-weighted portfolio of US stocks based on a boosted tree-based signal generates on average an excess return of 3.1% per annum, compared to a simple multifactor portfolio. We demonstrat...

Nowadays commodity investing is facing a tremendous interest from all kinds of investors, the surging amount invested in commodity related indices being one of the manifestations of this phenomenon. Due to their historical de-correlation with conventional securities and their hedging properties against inflation, commodity investment is being perce...

This paper examines the forecasting performance of GARCH’s models used with agricultural commodities data. We compare different possible sources of forecasting improvement, using various statistical distributions and models. We have chosen to confine our analysis on four indices which are the cocoa LIFFE continuous futures, the cocoa NYBOT continuo...