Search results

We provide a new characterization of the equality of two positive-definite matrices A and B, and we use this to propose several new computationally convenient statistical tests for the equality of two unknown positive-definite matrices. Our primary focus is on testing the information matrix equality (e.g. White, 1982, 1994). We characterize the asymptotic behavior of our new trace-determinant information matrix test statistics under the null and the alternative and investigate their finite-sample performance for a variety of models: linear regression, exponential duration, probit, and Tobit. The parametric bootstrap suggested by Horowitz (1994) delivers critical values that provide admirable level behavior, even in samples as small as n = 50. Our new tests often have better power than the parametric-bootstrap version of the traditional IMT; when they do not, they nevertheless perform respectably.

Most economic models in essence specify the mean of some explained variables, conditional on a number of explanatory variables. Since the publication of White’s (1982) Econometrica paper, a vast literature has been devoted to the quasi- or pseudo-maximum likelihood estimator (QMLE or PMLE). Among others, it was shown that QMLE of a density from the linear exponential family (LEF) provides a consistent estimate of the true parameters of the conditional mean, despite misspecification of other aspects of the conditional distribution. In this paper, we first show that it is not the case when the weighting matrix of the density and the mean parameter vector are functionally related. A prominent example is an autoregressive moving-average (ARMA) model with generalized autoregressive conditional heteroscedasticity (GARCH) error. As a result, the mean specification test is not readily modified as heteroscedasticity insensitive. However, correct specification of the conditional variance adds conditional moment conditions for estimating the parameters in conditional mean. Based on the recent literature of efficient instrumental variables estimator (IVE) or generalized method of moments (GMM), we propose an estimator which is modified upon the QMLE of a density from the quadratic exponential family (QEF). Moreover, GARCH-M is also allowed. We thus document a detailed comparison between the quadratic exponential QMLE with IVE. The asymptotic variance of this modified QMLE attains the lower bound for minimax risk.

In the typical study comparing the evolution of economic inequality among different territorial units, an inequality indicator is chosen, and its value is calculated from sample data. Thus, the problem turns out to be the selection of the inequality indicator.

This paper shows that there is no need for a selection of a single inequality indicator. A whole set of inequality indicators are considered and calculated for the European Countries, using income data from European Community Household Panel (ECHP). The information they provide is then collapsed into a composite inequality indicator, through an adaptation of Principal Component Analysis (PCA). We analyze the conditions needed to make longitudinal comparisons possible. Results obtained with this composite indicator are used to compare and analyze the trends in economic inequality in the EU Countries.

This study aims to investigate the impact of the macroeconomic environment on South African industrial sector returns.

Using standardized coefficients derived from time-series factor models, the authors quantify the impact of macroeconomic influences on industrial sector returns. The authors analyze the structure of the resultant residual correlation matrices to establish the level of factor omission and apply a factor analytic augmentation to arrive at a specification that is free of omitted common factors.

The authors find that global influences are the most important drivers of returns and that industrial sectors are highly integrated with the global economy. The authors show that specifications that comprise only macroeconomic factors and proxies for omitted factors in the form of residual market factors are likely to be underspecified. This study demonstrates that a factor analytic augmentation is an effective approach to ensuring an adequately specified model.

The findings have a number of implications that are of interest to investors, econometricians and researchers. While the study focusses on a single market, the South African stock market, as represented by the Johannesburg Stock Exchange (JSE), it is a highly developed and globally integrated market. In terms of market capitalization, it exceeds the Madrid Stock Exchange, the Taiwan Stock Exchange and the BM&F Bovespa. Yet, a limited number of studies investigate the macroeconomic drivers of the South African stock market.

Investors should be aware that while the South African domestic environment, especially political risk, has an impact on returns, global influences are the greatest determinants of returns. No industrial sectors are insulated from global influences and this limits the potential for diversification. This study suggests an alternative set of macroeconomic factors that may be used in further analysis and asset pricing studies. From an econometric perspective, this study demonstrates the usefulness of a factor analytic augmentation as a solution to factor omission in models that use macroeconomic factors to proxy for systematic influences that describe asset prices.

The contribution lies in providing insight into a large and well-developed yet understudied financial market, the South African stock market. This study considers a much broader set of macroeconomic factors than prior studies. A methodological contribution is made by estimating and interpreting standardized coefficients to discriminate between the impact of domestically and internationally driven factors. This study shows that should coefficients not be standardized, inferences relating to the relative importance of factors will differ. Finally, the authors unify an approach of using pre-specified factors with a factor analytic approach to address factor omission and to ensure a valid and readily interpretable specification.

Situational involvement (SI) and enduring involvement (EI) are important predictors of spectator sports tourist behaviours. For this study, onsite and web surveys were utilised to help understand how SI and EI levels, with both event and destination, may vary according to the primary and secondary trip purpose of a spectator sports tourist. Results revealed differences between the two groups only within certain aspects of SI and EI with the destination.

The purpose of this paper is to explore the major determinants that influence the management teachers to practice management consulting. The second objective of this research is to understand how the experience in management consultancy leads to value addition in their class room teaching.

To address the first research objective, focus group discussions were conducted with management teachers practicing consultancy. These results were used to generate items for the questionnaire. Factor analysis performed on the data revealed six determinants influencing management teachers to engage in consulting activity. To address the second research objective, focus group discussions with MBA graduates were used to comprehend how teachers with management consulting experience enrich the pedagogy.

The major findings of the study suggest that the determinants influencing management teachers to practice consulting are: improving competencies, furthering professional advancement, accruing strategic and financial benefit, enabling holistic development. Through study 2, the authors found out that management teachers add value in pedagogy by forging corporate world connection through real-time examples, enable critical thinking by breaking established paradigms, effective classroom delivery through storytelling, etc., and lending student support by assuming a mentor’s role.

This study found that faculty consulting reduces the perceived gap between the industry and academia and it also leads to effective class room teaching.

The study is the first attempt to empirically test the determinants influencing management teachers to practice consultancy services and qualitatively assess how the consultancy experience enriches the in-class performance.

The usefulness of ex‐post data as a proxy for ex‐ante returns in the portfolio problem rests on the stability of the co‐movement between returns. Yet despite its importance, this issue has not received sufficient examination in the financial literature, particularly in the direct real estate market. This study aims to address this issue.

To examine the temporal stability of covariance and correlation matrices and individual correlation coefficients this paper uses the Box M tests and the methodology of Shaked using monthly real estate data in the UK over the period 1987 to 2002 and four investment horizons.

The Box M tests reveal that the covariance and correlation matrices both display temporal instability. This suggests that the returns between real estate returns are unstable over time and so provide poor estimates in the ex‐ante modelling process. The analysis also indicates that the covariance matrices are less stable than the corresponding correlation matrices. Nonetheless, when we tested the stability of individual correlation coefficients using the methodology of Shaked we find that stability increases consistently and substantially with the lengthening of the investment horizon and holding period.

Thus, for all practical purposes the pair‐wise correlation between real estate returns can be considered nearly stationary in the long run. This implies that investors can use ex‐post data as a proxy for ex‐ante data in portfolio models especially if longer investment horizons are used to estimate the parameters.

This study is the first to examine temporal co‐movements between UK real estate returns in a portfolio context over different investment horizons.

As a class of stochastic hybrid systems, Markovian jump systems have been extensively studied in the past decades. In light of some results obtained on this topic. The purpose of this paper is to investigate the stability problems for delayed Markovian jump systems.

The time‐varying‐delays considered in this paper are switched synchronously with system mode. Based on stochastic Lyapunov theory, the delay‐dependent stability conditions are developed by using some linear matrix inequality techniques. To obtain better stability criteria, the different Lyapunov‐Krasovskii functional is chosen and an important inequality is introduced.

Numerical examples show that the resulting criteria in this paper have advantages over some previous ones in that they involve fewer matrix variables, but have less conservatism. Furthermore, they only involve the matrix variables appeared in the Lyapunov functional. Therefore, there are no additional matrix variables coupled with the system matrices, which will be easier to investigate the synthesis problems for the underlying systems and save much computation.

The introduced approach is more efficient to investigate the stability for Markovian jump systems with mode‐dependent time‐varying‐delays.

In this paper we talk about complex matrix quaternions (biquaternions) and we deal with some abstract methods in mathematical complex matrix analysis.

We introduce and investigate the complex space HC consisting of all 2 × 2 complex matrices of the form ξ=z1+iw1z2+iw2−z‾2−iw‾2z‾1+iw‾1, (z1,w1,z2,w2)∈C4.

We develop on HC a new matrix holomorphic structure for which we provide the fundamental operational calculus properties.

We give sufficient and necessary conditions in terms of Cauchy–Riemann type quaternionic differential equations for holomorphicity of a function of one complex matrix variable ξ∈HC. In particular, we show that we have a lot of holomorphic functions of one matrix quaternion variable.

To date, the literature on quantile regression and least absolute deviation regression has assumed either explicitly or implicitly that the conditional quantile regression model is correctly specified. When the model is misspecified, confidence intervals and hypothesis tests based on the conventional covariance matrix are invalid. Although misspecification is a generic phenomenon and correct specification is rare in reality, there has to date been no theory proposed for inference when a conditional quantile model may be misspecified. In this paper, we allow for possible misspecification of a linear conditional quantile regression model. We obtain consistency of the quantile estimator for certain “pseudo-true” parameter values and asymptotic normality of the quantile estimator when the model is misspecified. In this case, the asymptotic covariance matrix has a novel form, not seen in earlier work, and we provide a consistent estimator of the asymptotic covariance matrix. We also propose a quick and simple test for conditional quantile misspecification based on the quantile residuals.