Search

We propose an ensemble learning methodology to forecast the future US GDP growth release. Our approach combines a Recurrent Neural Network (RNN) with a Dynamic Factor model accounting for time-variation in the mean with a Generalized Autoregressive Score (DFM-GAS). We show how this combination improves forecasts in the aftermath of the 2008-09 global financial crisis, as well as in the latest COVID-19 recession, by reducing the root mean squared error for the short-term forecast horizon. Thus, we evaluate the predictive ability of each component of the ensemble by considering variations in the mean, as the latter are potentially caused by recessions affecting the economy. For our scope, we employ a set of predictors encompassing a wide range of variables measured at different time frequencies. Thus, we provide dynamic coefficients for predictors after an interpretable machine learning routine to assess how the model reflects the evolution of the business cycle.

This paper analyses the impact of the Italian Startup Act which entered into force in December 2012. This public policy provides a unique bundle of benefits, such as tax incentives, public loan guarantees, and a more flexible labor law, for firms registered as “innovative startups” in Italy. This legislation has been implemented by the Italian government to increase innovativeness of small and young enterprises by facilitating improved access to (external) capital and (highskilled) labor. Consequently, the goal of our evaluation is to assess the impact of the policy on equity, debt and employment. Using various conditional difference-in-difference models, we find that the Italian startup policy has met its primary objectives. The econometric results strongly suggest that firms operating under this program are more successful in obtaining equity and debt capital and they also hire more employees because of the program participation.

In the current COVID-19 pandemic, we witness sharp fluctuations of biopharmaceutical equity prices as a response to innovation related news. On 23.04.2020, the stock price of Biotech firm Gilead Sciences fell by as much as 9% after rumors suggested that remdesivir, its experimental COVID-19 treatment, showed no benefit. One week later, official clinical trial results suggesting the opposite have led to a rebounce of 11% in equity value. Jumps in equity prices of this amplitude are frequent in the biopharmaceutical industry, yet we lack knowledge about the underlying forces that drive them. In this paper we investigate the impact of product innovation on firm equity value by linking venture specific characteristics to stock market reactions using a large data set of the biopharmaceutical industry. We find that market reactions increase in the portfolio importance of the product under development, measured by product specific net present value data, and diminish in its success probability, estimated by a combination of supervised learning methods. Our results aid corporations and investors to evaluate the financial consequences of innovation related news.

In the field of risk management, scholars began to bring together the quantitative methodologies with the banking management issues about 30 years ago, with a special focus on market, credit and operational risks. After the systemic eff ects of banks defaults during the recent fi nancial crisis, and despite a huge amount of literature in the last years concerning the systemic risk, no standard methodologies have been set up to now. Even the new Basel 3 regulation has adopted a heuristic indicator-based approach, quite far from an e ffective quantitative tool. In this paper, we refer to the di fferent pieces of the puzzle: defi nition of systemic risk, a set of coherent and useful measures, the computability of these measures, the data set structure. In this challenging fi eld, we aim to build a comprehensive picture of the state of the art, to illustrate the open issues, and to outline some paths for a more successful future research. This work appropriately integrates other useful surveys and it is directed to both academic researchers and practitioners.

textabstractThis paper examines how a radical technological innovation affects alliance formation of firms and subsequent network structures. We use longitudinal data of interfirm R&D collaborations in the biopharmaceutical industry in which a new technological regime is established. Our findings suggest that it requires radical technological change for firms to leave their embedded path of existing alliances and form new alliances with new partners. While new partners are mostly found through the firms’ existing network, we provide some insight into distant link formation with unknown partners, which contributes to our understanding of how ‘small-worlds’ might emerge.