This paper proposes a new approach to estimating the minimum variance hedge ratio (MVHR) based on the wild bootstrap and evaluates the approach using a spectrum of conservative to aggressive alternative hedging strategies associated with the percentiles of the MVHR’s bootstrap distribution. This approach is suggested to be more informative and effective relative to the conventional method of hedging solely based on a single-point estimate. Furthermore, the percentile-based MVHRs are robust to influential outliers, non-normality, and unknown forms of heteroskedasticity. The bootstrap percentile-based hedging strategies’ effectiveness is compared with those from the naïve method and the asymmetric DCC-GARCH model for a range of financial assets and commodities. The bootstrap percentile-based hedging technique is identified to outperform its alternatives in terms of hedging effectiveness, downside risk, and return variability, suggesting its superiority to other methods in both the literature and in practice.

Download This Paper Open PDF in Browser Add Paper to My Library Share: Permalink Using these links will ensure access to this page indefinitely Copy URL Copy DOI

This paper proposes a test for the signal-to-noise ratio applicable to a range of significance tests and model diagnostics in a linear regression model. It is particularly useful when sample size is large or massive, where, as a consequence, conventional tests frequently lead to inappropriate rejection of the null hypothesis. The test is conducted in the context of the traditional F -test, with its critical values increasing with sample size. It maintains desirable size properties under a large or massive sample size, when the null hypothesis is violated by a practically negligible margin. The test is widely applicable to many empirical studies in business and management.

We conduct the extreme bounds analysis (EBA) to evaluate the robustness or fragility of a range of stock market anomalies, using U.S. daily data from 1960 to 2023. The EBA is a large-scale sensitivity analysis, able to isolate the effects of potential data-mining or p-hacking under model uncertainty. The anomalies covered include the effects of Halloween, sports event, seasonal affective disorder, weather, political cycle, daylight saving, and lunar phase. We find that the empirical evidence for the anomalies is highly fragile, in terms of effect size estimates and their statistical significance.

This paper evaluates the predictability of monthly stock return using out-of-sample interval forecasts. Past studies exclusively use point forecasts, which are of limited value since they carry no information about intrinsic predictive uncertainty. We compare the empirical performance of alternative interval forecasts for stock return generated from a naïve model, univariate autoregressive model, and multivariate model (predictive regression and VAR), using U.S. data from 1926. It is found that neither univariate nor multivariate interval forecasts outperform naïve forecasts. This strongly suggests that the U.S. stock market has been informationally efficient in the weak form as well as in the semistrong form.

Unmanned Aerial Vehicles (UAVs) find increasing use in mission critical tasks both in civilian and military operations. Most UAVs rely on Inertial Measurement Units (IMUs) to calculate vehicle attitude and track vehicle position. Therefore, an incorrect IMU reading can cause a vehicle to destabilize, and possibly even crash. In this paper, we describe how a strategic adversary might be able to introduce spurious IMU values that can deviate a vehicle from its mission-specified path while at the same time evade customary anomaly detection mechanisms, thereby effectively perpetuating a "stealthy attack" on the system. We explore the feasibility of a Deep Neural Network (DNN) that uses a vehicle's state information to calculate the applicable IMU values to perpetrate such an attack. The eventual goal is to cause a vehicle to perturb enough from its mission parameters to compromise mission reliability, while, from the operator's perspective, the vehicle still appears to be operating normally.

We decompose exchange rate exposure into systematic and partial parts. The former is the product of the exposure of the market portfolio and a firm’s market beta, reflecting the risk of the exchange rate to a macroeconomy. The latter is the residual one that most previous studies have examined. Using Japanese data, we find that Japanese firms are systematically exposed to the exchange rate from the beginning of 2000. We also highlight the timely yen-selling intervention by the Bank of Japan when the firms are systematically exposed. However, we find that, even when most Japanese firms are significantly exposed to the exchange rate, the partial exposure can seriously underestimate the full extent of the exchange rate exposure.

This paper evaluates the predictability of monthly stock return using out-of-sample (multi-step ahead and dynamic) prediction intervals. Past studies have exclusively used point forecasts, which are of limited value since they carry no information about the intrinsic predictive uncertainty associated. We compare empirical performances of alternative prediction intervals for stock return generated from a naive model, univariate autoregressive model, and multivariate model (predictive regression and VAR), using the U.S. data from 1926. For evaluation free from structural change and data snooping bias, we adopt moving sub-sample windows of different lengths. It is found that the naive model often provides the most informative prediction intervals, outperforming those generated from the univariate model as well as those from the multivariate models that incorporate a range of economic and financial predictors. This strongly suggests that the U.S. stock market has been informationally efficient in the weak-form as well as in the semi-strong form, subject to the information set considered in this study.

TROPOspheric Monitoring Instrument (TROPOMI) on board the Sentinel 5 P instrument was launched into space in October 2017. Meanwhile the analysed data from several research institutes offer a variety of trace gases columns, including different tropospheric ozone data set. One based on the Convective Cloud Differential algorithm is an official TROPOMI data product. The algorithm in an improvement of the algorithm developed for the ESA ozone CCI project. It uses the offline (OFFL) ozone total columns and the cloud data sets. The data agree well with data from other satellite missions, i.e. GOME-2 or OMI. Therefore they will extent the time series started in 1995 with GOME / ERS2 developed during the CCI project. However despite the long time series of the CCD data set, the CCD algorithm can only be applied within the tropics (20°S -20°N). A completely different approach uses data assimilation to constrain the stratospheric ozone and subtract the stratosphere from the total column. This offers the possibility to study tropospheric ozone on a global scale. We developed an algorithm for the scientific product that will be presented as well. Ozone observation from the microwave limb sounder (MLS) on the AURA satellite constrain the ozone distribution in the BASCOE (Belgian Assimilation System for Chemical ObsErvations) assimilation model. The BASCOE ozone concentrations are integrated above the tropopause to calculate the stratospheric column. Interpolated stratospheric columns are subtracted from the TROPOMI total ozone columns. The tropospheric ozone data agree reasonable well with comparable data from NASA using OMPS total columns, and within the tropics also with the CCD data. The algorithm was extend to be used also for other total ozone column observations like from GOME-2, OMI. But also geostationary observers like GEMS can in principal be used and might give intersting insight in the daily cycle of tropospheric ozone.

Highly dynamic mobile ad-hoc networks (MANETs) are continuing to serve as one of the most challenging environments to develop and deploy robust, efficient, and scalable routing protocols. In this paper, we present DeepCQ+ routing which, in a novel manner, integrates emerging multi-agent deep reinforcement learning (MADRL) techniques into existing Q-learning-based routing protocols and their variants, and achieves persistently higher performance across a wide range of MANET configurations while training only on a limited range of network parameters and conditions. Quantitatively, DeepCQ+ shows consistently higher end-to-end throughput with lower overhead compared to its Q-learning-based counterparts with the overall gain of 10-15% in its efficiency. Qualitatively and more significantly, DeepCQ+ maintains remarkably similar performance gains under many scenarios that it was not trained for in terms of network sizes, mobility conditions, and traffic dynamics. To the best of our knowledge, this is the first successful demonstration of MADRL for the MANET routing problem that achieves and maintains a high degree of scalability and robustness even in the environments that are outside the trained range of scenarios. This implies that the proposed hybrid design approach of DeepCQ+ that combines MADRL and Q-learning significantly increases its practicality and explainability because the real-world MANET environment will likely vary outside the trained range of MANET scenarios.

As the guest editors of this Special Issue, we feel proud and grateful to write the editorial note of this issue, which consists of seven high-quality research papers [...]

This paper proposes a test for the signal-to-noise ratio applicable to a range of significance tests and model diagnostics in a linear regression. It is particularly useful under a large or massive sample size, where a conventional test frequently rejects an economically negligible deviation from the null hypothesis. The test is conducted in the context of the traditional $F$-test, with its critical values increasing with sample size. It maintains desirable size properties under a large or massive sample size, when the null hypothesis is violated by a practically negligible margin.

The level of significance should be chosen with careful consideration of the key factors such as the sample size, power of the test, and expected losses from Type I and II errors. While the conventional levels may still serve as practical benchmarks, they should not be adopted mindlessly and mechanically for every application.

Highly dynamic mobile ad-hoc networks (MANETs) remain as one of the most challenging environments to develop and deploy robust, efficient, and scalable routing protocols. In this paper, we present DeepCQ+ routing protocol which, in a novel manner, integrates emerging multi-agent deep reinforcement learning (MADRL) techniques into existing Q-learning-based routing protocols and their variants, and achieves persistently higher performance across a wide range of topology and mobility configurations. While keeping the overall protocol structure of the Q-learning-based routing protocols, DeepCQ+ replaces statically configured parameterized thresholds and hand-written rules with carefully designed MADRL agents such that no configuration of such parameters is required a priori. Extensive simulation shows that DeepCQ+ yields significantly increased end-to-end throughput with lower overhead and no apparent degradation of end-to-end delays (hop counts) compared to its Q-learning-based counterparts. Qualitatively, and perhaps more significantly, DeepCQ+ maintains remarkably similar performance gains under many scenarios that it was not trained for in terms of network sizes, mobility conditions, and traffic dynamics. To the best of our knowledge, this is the first successful application of the MADRL framework for the MANET routing problem that demonstrates a high degree of scalability and robustness even under the environments that are outside the trained range of scenarios. This implies that our MARL-based DeepCQ+ design solution significantly improves the performance of Q-learning-based CQ+ baseline approach for comparison and increases its practicality and explainability because the real-world MANET environment will likely vary outside the trained range of MANET scenarios. Additional techniques to further increase the gains in performance and scalability are discussed.

In light of the recent statements on the p -value criterion made by the American Statistical Association (ASA), it is now clear that that the current paradigm of statistical significance and its decision rule should be modified, in many fields of science including the business disciplines. We need a new or modified paradigm for “thoughtful, open and modest” research. This paper explains why we should adopt the ASA recommendations, and proposes a range of possible alternatives that may be included in a new paradigm for statistical research. As an application, the effect of seasonal affective disorder on stock return is re-evaluated.

The Geostationary Environment Monitoring Spectrometer (GEMS) is scheduled for launch in February 2020 to monitor air quality (AQ) at an unprecedented spatial and temporal resolution from a geostationary Earth orbit (GEO) for the first time. With the development of UV–visible spectrometers at sub-nm spectral resolution and sophisticated retrieval algorithms, estimates of the column amounts of atmospheric pollutants (O3, NO2, SO2, HCHO, CHOCHO, and aerosols) can be obtained. To date, all the UV–visible satellite missions monitoring air quality have been in low Earth orbit (LEO), allowing one to two observations per day. With UV–visible instruments on GEO platforms, the diurnal variations of these pollutants can now be determined. Details of the GEMS mission are presented, including instrumentation, scientific algorithms, predicted performance, and applications for air quality forecasts through data assimilation. GEMS will be on board the Geostationary Korea Multi-Purpose Satellite 2 (GEO-KOMPSAT-2) satellite series, which also hosts the Advanced Meteorological Imager (AMI) and Geostationary Ocean Color Imager 2 (GOCI-2). These three instruments will provide synergistic science products to better understand air quality, meteorology, the long-range transport of air pollutants, emission source distributions, and chemical processes. Faster sampling rates at higher spatial resolution will increase the probability of finding cloud-free pixels, leading to more observations of aerosols and trace gases than is possible from LEO. GEMS will be joined by NASA’s Tropospheric Emissions: Monitoring of Pollution (TEMPO) and ESA’s Sentinel-4 to form a GEO AQ satellite constellation in early 2020s, coordinated by the Committee on Earth Observation Satellites (CEOS).

This article is a primer for a decision-theoretic approach to hypothesis testing for students and teachers of basic statistics. Using three examples at an introductory level, this article demonstrates how decision-theoretic hypothesis testing can be taught to the students of basic statistics. It also demonstrates that students and researchers can make more sensible and unambiguous decisions under uncertainty by employing this particular approach. The examples are illustrated using R and its package "OptSig."

Serious concerns have been raised that false positive findings are widespread in empirical research in business disciplines. This is largely because researchers almost exclusively adopt the 'p-value less than 0.05' criterion for statistical significance; and they are often not fully aware of large-sample biases which can potentially mislead their research outcomes. This paper proposes that a statistical toolbox (rather than a single hammer) be used in empirical research, which offers researchers a range of statistical instruments, including a range of alternatives to the p-value criterion such as the Bayesian methods, optimal significance level, sample size selection, equivalence testing and exploratory data analyses. It is found that the positive results obtained under the p-value criterion cannot stand, when the toolbox is applied to three notable studies in finance.

Software-Defined Network (SDN) based battlefield network consists of network providers, such as Satellite Communication (SATCOM) systems and Unmanned Aerial Vehicles (UAVs), and users as battlefield entities. Battlefield entities, commonly equipped with multiple terminals, can obtain multiple communication links for multiple applications and system robustness. To better utilize the network resources and improve the communication performance for such network, we developed an integrated Software-Defined Network emulation testbed to support various tactical scenarios. In our continuing research, our proposed framework couples SDN and Multi-Path TCP (MPTCP) with a smart agent for real-time traffic optimization based on the Flow Deviation Method (FDM). In this paper, we enhance the testbed with greater control over traffic generation, ampler visualization options, dynamic link management to simulate network events, and support of larger topologies. Iterative experimentation, regression testing, and comparative analysis attest to the functionality and scalability of our integrated testbed, in support of further research and study on battlefield network and other tactical network environments.

Lambertian cloud model (Lambertian Cloud Model) is the simplified cloud model which is used to effectively retrieve the vertical ozone distribution of the atmosphere where the clouds exist. By using the Lambertian cloud model, the optical characteristics of clouds required for radiative transfer simulation are parametrized by Optical Centroid Cloud Pressure (OCCP) and Effective Cloud Fraction (ECF), and the accuracy of each parameter greatly affects the radiation simulation accuracy. However, it is very difficult to generalize the vertical ozone error due to the OCCP error because it varies depending on the radiation environment and algorithm setting. In addition, it is also difficult to analyze the effect of OCCP error because it is mixed with other errors that occur in the vertical ozone calculation process. This study analyzed the ozone retrieval error due to OCCP error using two methods. First, we simulated the impact of OCCP error on ozone retrieval based on Optimal Estimation. Using LIDORT radiation model, the radiation error due to the OCCP error is calculated. In order to convert the radiation error to the ozone calculation error, the radiation error is assigned to the conversion equation of the optimal estimation method. The results show that when the OCCP error occurs by 100 hPa, the total ozone is overestimated by 2.7%. Second, a case analysis is carried out to find the ozone retrieval error due to OCCP error. For the case analysis, the ozone retrieval error is simulated assuming OCCP error and compared with the ozone error in the case of PROFOZ 2005-2006, an OMI ozone profile product. In order to define the ozone error in the case, we assumed an ideal assumption. Considering albedo, and the horizontal change of ozone for satisfying the assumption, the 49 cases are selected. As a result, 27 out of 49 cases (about 55%) showed a correlation of 0.5 or more. This result show that the error of OCCP has a significant influence on the accuracy of ozone profile calculation.