Nonlinear dynamical structural equation models

Structural equation models (SEMs) have been widely used in behavioural, educational, medical and socio-psychological research for exploring and confirming relations among observed and latent variables. In the existing SEMs, the unknown coefficients in the measurement and structural equations are assumed to be constant with respect to time. This assumption does not always hold, as the relation among the observed and latent variables varies with time for some situations. In this paper, we propose nonlinear dynamical structural equation models to cope with these situations, and explore the nonlinear dynamic of the relation between the variables involved. A local maximum likelihood-based estimation procedure is proposed. We investigate a bootstrap resampling-based test for the hypothesis that the coefficient is constant with respect to time, as well as confidence bands for the unknown coefficients. Intensive simulation studies are conducted to show the empirical performance of the proposed estimation procedure, hypothesis test statistic and confidence band. Finally, a real example in relation to the stock market of Hong Kong is presented to demonstrate the proposed methodologies.     

Testing for structural breaks in return-based style regression models

Financial Markets and Portfolio Management - It is important for investors to know not only the style of a fund manager in which they are interested, but also whether this style is constant or...     

A structural equation modelling of the antecedents of relationship quality in the Malaysia banking sector

The intense competition among firms in the new global environment has made it inevitable for firms to seek ways to create and maintain quality relationship with customers, and Malaysia banks are no exception. However, little is understood from empirical viewpoint about the antecedents of relationship quality specifically, the actual influence of overall customer satisfaction and its indicators. Based on data collected from 220 customers of 15 retail banks in Malaysia and analysed using the structural equation modelling technique, findings are that overall customer satisfaction is a key determinant of relationship quality. The indicators of customer satisfaction include trust, commitment, communication, service quality, service satisfaction and conflict handling. These results are not confounded by any service differential among the participating banks. Important theoretical and managerial implications of the findings are discussed.     

Asymmetric volatility in the cryptocurrency market: New evidence from models with structural breaks

Previous literature shows that major cryptocurrencies exhibit inverse asymmetric volatility: positive shocks increase price volatility more than negative ones. In this study, we revisit the asymmetric volatility dynamics of major cryptocurrencies using asymmetric GARCH models that incorporate endogenously detected structural breaks. Our results show that after incorporating structural breaks, volatility persistence decreases and asymmetric volatility increases for all cryptocurrencies in this study. Thus, prior research that ignores structural breaks underestimates the impact of unexpected news on price volatility in cryptocurrency markets. We also present important economic implications of our results: ignoring structural breaks adversely affects the hedging strategies, derivatives valuations, and risk exposure measurement of investors in cryptocurrency markets.     

Measuring risk in the mining sector with ARCH models with important observations on sample size

This paper uses the GARCH technique to estimate time-varying individual firm risk measures for the mining sector. In general, the mining industry is riskier than the market with estimated betas greater than one. The results also show that the level of risk in the industry is quite volatile and can be divided into three distinct time periods according to the magnitude and variability of the firm's betas. Additionally, the number of observations in a sample has a strong relationship to the detection and estimation of an ARCH effect in the data, which cannot be explained by firm financial size.     

Instability of sunspot equilibria in real business cycle models under adaptive learning

We provide conditions under which a general, reduced-form class of real business cycle (RBC) models has rational expectations equilibria that are both indeterminate and stable under adaptive learning. Indeterminacy of equilibrium allows for the possibility that non-fundamental “sunspot” variable realizations can be used to drive the model, and several researchers have offered calibrated structural models where sunspot shocks play such a role. However, we show that the structural restrictions researchers have adopted lead to reduced-form systems that are always unstable under adaptive learning dynamics, thus calling into question the plausibility of these sunspot-driven RBC models.