This study investigates the interconnection between five implied volatility indices representative of different financial markets during period 1 August 2008–29 December 2017. To this end, we first perform a static and dynamic analysis to measure total connectedness in entire (the system-wide approach) using framework recently proposed by Diebold Yilmaz. Second, make use evaluate both net directional for each market all pairwise connectedness. Our results suggest that 38.99%, variance...

We propose a methodology for estimating the evolution of epidemiological parameters SIRD model (acronym Susceptible, Infected, Recovered and Deceased individuals) which allows to evaluate sanitary measures taken by government, COVID-19 in Spanish outbreak. In our only information required these is time series deceased people; due number asymptomatic people produced COVID-19, it not possible know actual infected at any given time. Therefore, among different that quantify pandemic we consider...

Abstract We present a system for combining the different types of predictions given by wide category mechanical trading rules through statistical learning methods (boosting, and several model averaging like Bayesian or simple methods). Statistical supply better out‐of‐sample results than most single moving average in NYSE Composite Index from January 1993 to December 2002. Moreover, using filter reduce frequency, filtered boosting produces technical strategy which, although it is not able...

In this paper, we examine the dynamic behaviour of US stock market due to subsequent impact COVID-19 outbreak and war in Ukraine. To that end, analyse daily data Dow Jones Industrial Average returns from 2 January 1900 31 October 2022. Firstly, identify past crisis episodes similar current situation. Then, compare volatility dynamics, variation-fluctuation correlation functions, with uncertainty indicators those induced by epidemic Russo-Ukrainian conflict. Our findings suggest consecutive...

Abstract We propose a new test to detect chaotic dynamics, based on the stability of largest Lyapunov exponent from different sample sizes. This is applied data used in single‐blind controlled competition tests for non‐linearity and chaos that were generated by Barnett et al. ( 1997 ), as well several other series. The results suggest particularly effective when compared stochastic alternatives (both linear non‐linear). For large sizes power one, although small it diminishes occasionally....

This article assesses the economic significance of non-linear predictability EMS exchange rates. To that end, and using daily data for nine currencies covering 1 January 1978–31 December 1994 period, it considers nearest- neighbour predictors, transforming their forecasts into a technical trading rule, whose profitability has been evaluated against traditional moving average rules, considering both interest rates transaction costs. The results suggest in most cases, rule based on predictor...

The purpose of this paper is to contribute the debate on relevance non-linear predictors high-frequency data in foreign exchange markets. To that end, we apply nearest-neighbour (NN) predictors, inspired by literature forecasting dynamical systems, exchange-rate series. performance univariate and multivariate versions such NN first evaluated from statistical point view, using a battery tests. Secondly, assess if are capable producing valuable economic signals results show potential...

In this paper we assess whether some simple forms of technical analysis can predict stock price movements in the Madrid Stock Exchange. To that end, use daily data for General Index Exchange, covering thirty-one-year period from January 1966-October 1997.

In this paper we investigate the profitability of non-linear trading rules based on nearest neighbour (NN) predictors. Applying investment strategy to New York Stock Exchange for 1997-2002 period, our results suggest that, taking into account transaction costs, NN-based rule is superior both a risk-adjusted buy-and-hold and linear ARIMA-based in terms returns all years studied, except 2000 2001. addition, produces higher net than those from simple strategy, 1997. Regarding other measures,...

This study employs different nonlinear models (smooth transition autoregressive (STAR), artificial neural networks (ANN) and nearest neighbours (NN)) to the predictability of one-step-ahead forecast returns for Ibex35 stock future index at a one year horizon. It is found that STAR, ANN NN beat random walk (RW) linear (AR) in out-of-sample statistical accuracy, also when economic criteria were used simple trading strategy including impact transaction costs on profits. Finally, overall results...

In this paper we investigate the profitability of non-linear trading rules based on nearest neighbor predictors. Applying investment strategy to New York Stock Exchange, our results suggest that, taking into account transaction costs, rule is superior a risk-adjusted buy-and-hold (both in terms returns and Sharpe ratios) for 1998 1999 periods upward trend. contrast, relatively "stable" market period 2000, found that both strategies generate equal returns, although yields higher ratio.

In this article, we analyse the co-movements of daily stock prices and government bond during last 25 years, in major Western markets, extending previous results to take into account impact current crisis. Our confirm that bonds are viewed as instruments for improving portfolio diversification periods high volatility falling market levels, which is when such most needed. The possibility using debt portfolios a means hedging times financial crisis became especially apparent crises 1997, 2001...

This paper examines whether short-term rental listings in the sharing accommodation market take account of risk their pricing. To do so, we estimate time-varying risks, and forecast price changes using daily supply-price time series. The empirical analysis was conducted data for Canary Islands period January 2016 to September 2021. following main results were obtained. First, individual face systematic risks that are lower than average listing return, but multi-unit hosts more sensitive...

New evidence is presented on the positive correlation between returns from technical trading rules and periods of central bank intervention. To that end, profitability a strategy based nearest-neighbour (nonlinear) predictors evaluated, which may be viewed as generalization graphical methods widely used in financial markets. Daily data US dollar/Deutschmark dollar/Japanese Yen covering 1 February 1982–31 December 1996 period are used. The results suggest exclusion days intervention implies...

In this paper we assess the economic significance of nonlinear predictability EMS exchange rates. To that end, and using daily data for nine currencies covering 1st January 1978-31st December 1994 period, consider nearest-neighbour predictors, transforming their forecasts into a technical trading rule, whose profitability has been evaluated against traditional moving average rules, considering both interest rates transaction costs. Our results suggest in most cases, rule based on predictor...

This study investigates the interconnection between five implied volatility indices representative of different financial markets during period August 1, 2008-September 9, 2015. To this end, we first perform a static and dynamic analysis to measure total connectedness in entire (the system-wide approach) using framework recently proposed by Diebold Yılmaz (2014). Second, make use evaluate both net directional for each market all pair-wise connectedness. Our results suggest that slightly more...

We propose a new test to detect chaotic dynamics, based on the stability of largest Lyapunov exponent from different sample sizes. This is applied data used in single-blind controlled competition tests for nonlinearity and chaos that were generated by Barnett et al. (1997), as well several series. The results suggest particularly effective when compared other stochastic alternatives (both linear nonlinear). size one large samples, although small sizes it diminishes below nominal two out...

In this paper we assess the economic significance of nonlinear predictability EMS exchange rates. To that end, and using daily data for nine currencies covering 1st January 1978-31st December 1994 period, consider nearest-neighbour predictors, transforming their forecasts into a technical trading rule, whose profitability has been evaluated against traditional (linear) moving average rules, considering both interest rates transaction costs. Our results suggest in most cases rule based on...