|
on Financial Markets |
|
Issue of 2020‒04‒20
twelve papers chosen by |
| By: | Scott R. Baker; Nicholas Bloom; Steven J. Davis; Kyle J. Kost; Marco C. Sammon; Tasaneeya Viratyosin |
| Abstract: | No previous infectious disease outbreak, including the Spanish Flu, has impacted the stock market as powerfully as the COVID-19 pandemic. We use text-based methods to develop this point with respect to large daily stock market moves back to 1900 and with respect to overall stock market volatility back to 1985. We also argue that policy responses to the COVID-19 pandemic provide the most compelling explanation for its unprecedented stock market imact. |
| JEL: | E44 G12 |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:26945&r=all |
| By: | Stefano Giglio; Matteo Maggiori; Johannes Stroebel; Stephen Utkus |
| Abstract: | We provide a data-driven analysis of how investor expectations about economic growth and stock market returns changed during the February-March 2020 stock market crash induced by the COVID-19 pandemic. We surveyed wealthy retail investors who are clients of Vanguard in mid-February 2020, around the all-time stock market high, and then again on March 11 and 12, after the stock market had collapsed by over 20%. The average investor turned more pessimistic about the short-run performance of both stock markets and the economy. Investors also perceived higher probability of both further extreme stock market declines and large declines in short-run real economic activity. In contrast, investors' expectations about the long run remained largely unchanged, and if anything improved. Disagreement among investors about economic and stock market outcomes also increased substantially. Our analysis is an input in both the design of the ongoing economic policy response and in further advancing economic theories. |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.01831&r=all |
| By: | Tomaso Aste |
| Abstract: | The multivariate conditional probability distribution quantifies the effects of a set of variables onto the statistical properties of another set of variables. In the study of systemic risk in the financial system, the multivariate conditional probability distribution can be used for stress-testing by quantifying the propagation of losses from a set of `stressing' variables to another set of `stressed' variables. Here it is described how to compute such conditional probability distributions for the vast family of multivariate elliptical distributions, which includes the multivariate Student-t and the multivariate Normal distributions. Simple measures of stress impact and systemic risk are also proposed. An application to the US equity market illustrates the potentials of this approach. |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.06420&r=all |
| By: | Mojtaba Nabipour; Pooyan Nayyeri; Hamed Jabani; Amir Mosavi |
| Abstract: | Prediction of stock groups' values has always been attractive and challenging for shareholders. This paper concentrates on the future prediction of stock market groups. Four groups named diversified financials, petroleum, non-metallic minerals and basic metals from Tehran stock exchange are chosen for experimental evaluations. Data are collected for the groups based on ten years of historical records. The values predictions are created for 1, 2, 5, 10, 15, 20 and 30 days in advance. The machine learning algorithms utilized for prediction of future values of stock market groups. We employed Decision Tree, Bagging, Random Forest, Adaptive Boosting (Adaboost), Gradient Boosting and eXtreme Gradient Boosting (XGBoost), and Artificial neural network (ANN), Recurrent Neural Network (RNN) and Long short-term memory (LSTM). Ten technical indicators are selected as the inputs into each of the prediction models. Finally, the result of predictions is presented for each technique based on three metrics. Among all the algorithms used in this paper, LSTM shows more accurate results with the highest model fitting ability. Also, for tree-based models, there is often an intense competition between Adaboost, Gradient Boosting, and XGBoost. |
| Date: | 2020–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.01497&r=all |
| By: | Sven Husmann; Antoniya Shivarova; Rick Steinert |
| Abstract: | The recent advancements in computational power and machine learning algorithms have led to vast improvements in manifold areas of research. Especially in finance, the application of machine learning enables researchers to gain new insights into well-studied areas. In our paper, we demonstrate that unsupervised machine learning algorithms can be used to visualize and classify company data in an economically meaningful and effective way. In particular, we implement the t-distributed stochastic neighbor embedding (t-SNE) algorithm due to its beneficial properties as a data-driven dimension reduction and visualization tool in combination with spectral clustering to perform company classification. The resulting groups can then be implemented by experts in the field for empirical analysis and optimal decision making. By providing an exemplary out-of-sample study within a portfolio optimization framework, we show that meaningful grouping of stock data improves the overall portfolio performance. We, therefore, introduce the t-SNE algorithm to the financial community as a valuable technique both for researchers and practitioners. |
| Date: | 2020–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.01496&r=all |
| By: | Philip Ndikum |
| Abstract: | This research paper explores the performance of Machine Learning (ML) algorithms and techniques that can be used for financial asset price forecasting. The prediction and forecasting of asset prices and returns remains one of the most challenging and exciting problems for quantitative finance and practitioners alike. The massive increase in data generated and captured in recent years presents an opportunity to leverage Machine Learning algorithms. This study directly compares and contrasts state-of-the-art implementations of modern Machine Learning algorithms on high performance computing (HPC) infrastructures versus the traditional and highly popular Capital Asset Pricing Model (CAPM) on U.S equities data. The implemented Machine Learning models - trained on time series data for an entire stock universe (in addition to exogenous macroeconomic variables) significantly outperform the CAPM on out-of-sample (OOS) test data. |
| Date: | 2020–03 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.01504&r=all |
| By: | Guihai Zhao |
| Abstract: | Some key features in the historical dynamics of U.S. Treasury bond yields—a trend in long-term yields, business cycle movements in short-term yields, and a level shift in yield spreads—pose serious challenges to existing equilibrium asset pricing models. This paper presents a new equilibrium model to jointly explain these key features. The trend is generated by learning from the stable components in GDP growth and inflation, which share similar patterns to the neutral rate of interest (R-star) and trend inflation (Pi-star) estimates in the literature. Cyclical movements in yields and spreads are mainly driven by learning from the transitory components in GDP growth and inflation. The less-frequent inverted yield curves observed after the 1990s are due to the recent secular stagnation and procyclical inflation expectation. |
| Keywords: | Asset Pricing; Financial markets; Interest rates |
| JEL: | E43 G12 |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocawp:20-14&r=all |
| By: | Rui Fan (Swansea University); Oleksandr Talavera (University of Birmingham); Vu Tran (University of Reading) |
| Abstract: | This paper examines whether social media information affects the price discovery process for cross-listed companies. Using over 29 million overnight tweets mentioning cross-listed companies, we investigate the role of social media for the linkage between the last periods of trading in the US markets and the first periods in the UK market. Our estimates suggest that the size and content of information flows in social networks support the price discovery process. The interactions between lagged US stock features and overnight tweets significantly affect stock returns and volatility of cross-listed stocks when the UK market opens. These effects weaken and disappear after one to three hours after the UK market opening. We also develop a profitable trading strategy based on overnight social media, and the profits remain economically significant after considering transaction costs. |
| Keywords: | Twitter, investor sentiment, cross-listed stocks, text classification, computational linguistics |
| JEL: | G12 G14 L86 |
| Date: | 2020–03 |
| URL: | http://d.repec.org/n?u=RePEc:bir:birmec:20-05&r=all |
| By: | Maria Elvira Mancino; Simone Scotti; Giacomo Toscano |
| Abstract: | We empirically investigate the functional link between the variance swap rate and the spot variance. Using S\&P500 data over the period 2006-2018, we find overwhelming empirical evidence supporting the affine link analytically found by Kallsen et al. (2011) in the context of exponentially affine stochastic volatility models. Tests on yearly subsamples suggest that exponentially mean-reverting variance models provide a good fit during periods of extreme volatility, while polynomial models, introduced in Cuchiero (2011), are suited for years characterized by more frequent price jumps. |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.04015&r=all |
| By: | F. N. M. de Sousa Filho; J. N. Silva; M. A. Bertella; E. Brigatti |
| Abstract: | In this paper, we explore some stylized facts in the Bitcoin market using the BTC-USD exchange rate time series of historical intraday data from 2013 to 2018. Despite Bitcoin presents some very peculiar idiosyncrasies, like the absence of macroeconomic fundamentals or connections with underlying asset or benchmark, a clear asymmetry between demand and supply and the presence of inefficiency in the form of very strong arbitrage opportunity, all these elements seem to be marginal in the definition of the structural statistical properties of this virtual financial asset, which result to be analogous to general individual stocks or indices. In contrast, we find some clear differences, compared to fiat money exchange rates time series, in the values of the linear autocorrelation and, more surprisingly, in the presence of the leverage effect. We also explore the dynamics of correlations, monitoring the shifts in the evolution of the Bitcoin market. This analysis is able to distinguish between two different regimes: a stochastic process with weaker memory signatures and closer to Gaussianity between the Mt. Gox incident and the late 2015, and a dynamics with relevant correlations and strong deviations from Gaussianity before and after this interval. |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.05870&r=all |
| By: | Guglielmo Maria Caporale; Luis A. Gil-Alana; Miguel Martin-Valmayor |
| Abstract: | This paper examines the stochastic behaviour of the realized betas within the one-factor CAPM for the six companies with the highest market capitalization included in the Spanish IBEX stock market index. Fractional integration methods are applied to estimate their degree of persistence at the daily, weekly and monthly frequency over the period 1 January 2000 – 15 November 2018 using 1, 3 and 5-year samples. On the whole, the results indicate that the realized betas are highly persistent and do not exhibit mean-reverting behaviour. However, the findings are rather sensitive to the choice of frequency and time span (number of observations). |
| Keywords: | realized beta, CAPM, persistence, mean reversion, long memory |
| JEL: | C22 G11 |
| Date: | 2020 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_8171&r=all |
| By: | Huimin Peng |
| Abstract: | We analyze actively managed mutual funds in China from 2005 to 2017. We develop performance measures for asset allocation and selection. We find that stock selection ability from holding-based model is positively correlated with selection ability estimated from Fama-French three-factor model, which is price-based regression model. We also find that industry allocation from holding-based model is positively correlated with timing ability estimated from price-based Treynor-Mazuy model most of the time. We conclude that most actively managed funds have positive stock selection ability but not asset allocation ability, which is due to the difficulty in predicting policy changes. |
| Date: | 2020–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2004.05322&r=all |