Search results

This paper is concerned with the problem of estimating the uncertainty associated with a variable in a finite population. The study of this problem leads to the following conclusion: The classical measure of uncertainty, Shannon's entropy, is not suitable for sampling from finite populations; nevertheless, by using the entropy of order ? = 2, proposed by Havrda and Charvat, one can define an unbiased estimator of the uncertainty associated with the variable in both, the sampling with replacement and the sampling without replacement. This conclusion will be illustrated by an example.

The purpose of this manuscript is to quantify exactly how much information and/or predictive content is contained in business sentiment surveys.

This paper uses techniques drawn from information theory econometrics, and more specifically the theory of information entropy, to characterize the predictive content of business sentiment surveys. The authors apply these techniques to publicly available information obtained from various editions of the Federal Reserve Bank of New York's Empire State Manufacturing Survey, one of the most popular business sentiment surveys conducted in the USA. Parametric and non‐parametric statistical analyses are used to examine differences in the quantity of predictive content across various questions in the survey.

The results suggest that business sentiment surveys contain a reasonably high degree of informative content. However, the amount of informative content varies considerably from question to question in the survey. Questions that are more general in nature and ask about current perceptions (as opposed to future expectations) contain more informative content.

Business sentiment surveys are a practical, low‐cost method to assess the current and expected future state of local and regional economies. However, the value of these surveys is questionable if they do not contain much information. This research finds that such surveys do contain a large amount of information, and are worth administering. However, specific types of survey items convey more information that others, which also suggests that business sentiments surveys can be further revised to maximize the amount of content gained from respondents.

The purpose of this paper is to detect the occurrence of anomaly and fault in a spacecraft, investigate various tendencies of telemetry parameters and evaluate the operation state of the spacecraft to monitor the health of the spacecraft.

This paper proposes a data-driven method (empirical mode decomposition-sample entropy-principal component analysis [EMD-SE-PCA]) for monitoring the health of the spacecraft, where EMD is used to decompose telemetry data and obtain the trend items, SE is utilised to calculate the sample entropies of trend items and extract the characteristic data and squared prediction error and statistic contribution rate are analysed using PCA to monitor the health of the spacecraft.

Experimental results indicate that the EMD-SE-PCA method could detect characteristic parameters that appear abnormally before the anomaly or fault occurring, could provide an abnormal early warning time before anomaly or fault appearing and summarise the contribution of each parameter more accurately than other fault detection methods.

The proposed EMD-SE-PCA method has high level of accuracy and efficiency. It can be used in monitoring the health of a spacecraft, detecting the anomaly and fault, avoiding them timely and efficiently. Also, the EMD-SE-PCA method could be further applied for monitoring the health of other equipment (e.g. attitude control and orbit control system) in spacecraft and satellites.

The paper provides a data-driven method EMD-SE-PCA to be applied in the field of practical health monitoring, which could discover the occurrence of anomaly or fault timely and efficiently and is very useful for spacecraft health diagnosis.

Considering the problem that the high recognition rate of deep learning requires the support of mass data, this study aims to propose an insulating fault identification method based on small data set convolutional neural network (CNN).

Because of the chaotic characteristics of partial discharge (PD) signals, the equivalent transformation of the PD signal of unit power frequency period is carried out by phase space reconstruction to derive the chaotic features. At the same time, geometric, fractal, entropy and time domain features are extracted to increase the volume of feature data. Finally, the combined features are constructed and imported into CNN to complete PD recognition.

The results of the case study show that the proposed method can realize the PD recognition of small data set and make up for the shortcomings of the methods based on CNN. Also, the 1-CNN built in this paper has better recognition performance for four typical insulation faults of cable accessories. The recognition performance is improved by 4.37% and 1.25%, respectively, compared with similar methods based on support vector machine and BPNN.

In this paper, a method of insulation fault recognition based on CNN with small data set is proposed, which can solve the difficulty to realize insulation fault recognition of cable accessories and deep data mining because of insufficient measure data.

This paper aims to investigate whether the choice for a Big4-affiliated local audit firm affects the capital structure of listed companies in Indonesia, a fast-growing emerging country that is characterized by high information asymmetry and low litigation risk. A unique characteristic of the Indonesian audit environment is that Big4 auditors can only enter the market indirectly through affiliation with a local audit firm.

A sample of Indonesian listed companies between 2008 and 2015 is used to investigate this relation using ordinary least squares (OLS). To address the concern that the choice for Big4-affiliated auditors might reflect client characteristics, the authors also perform OLS on a matched sample, using both propensity-score and entropy-balance matching.

Across all three samples, the authors document that companies audited by a Big4-affiliated local audit firm display lower debt ratios. The authors find no such effect for affiliation with second-tier audit firms. Surprisingly, they find that the negative effect of Big4 affiliation is increasing in client size.

This study provides evidence of the consequences of hiring Big4 auditors on the perceived information asymmetry by financial markets under extreme conditions: in an environment characterized by low litigation risk and where Big4 auditors can operate only indirectly through affiliation.

The results of this study are of interest to policymakers, managers and financial stakeholders in emerging countries where external financing is important yet difficult to obtain because of severe information asymmetry. Hiring a Big4 auditor, even only through affiliation, might reduce perceived information asymmetry and increase the access to external equity financing.

To the best of the authors’ knowledge, this study is the first to provide evidence of the effect of Big4 auditors on their clients’ capital structure when they can operate only indirectly through affiliation with a local auditor.

Depression is a mental health problem characterized by a persistent sense of sadness and loss of interest. EEG signals are regarded as the most appropriate instruments for diagnosing depression because they reflect the operating status of the human brain. The purpose of this study is the early detection of depression among people using EEG signals.

(i) Artifacts are removed by filtering and linear and non-linear features are extracted; (ii) feature scaling is done using a standard scalar while principal component analysis (PCA) is used for feature reduction; (iii) the linear, non-linear and combination of both (only for those whose accuracy is highest) are taken for further analysis where some ML and DL classifiers are applied for the classification of depression; and (iv) in this study, total 15 distinct ML and DL methods, including KNN, SVM, bagging SVM, RF, GB, Extreme Gradient Boosting, MNB, Adaboost, Bagging RF, BootAgg, Gaussian NB, RNN, 1DCNN, RBFNN and LSTM, that have been effectively utilized as classifiers to handle a variety of real-world issues.

1. Among all, alpha, alpha asymmetry, gamma and gamma asymmetry give the best results in linear features, while RWE, DFA, CD and AE give the best results in non-linear feature. 2. In the linear features, gamma and alpha asymmetry have given 99.98% accuracy for Bagging RF, while gamma asymmetry has given 99.98% accuracy for BootAgg. 3. For non-linear features, it has been shown 99.84% of accuracy for RWE and DFA in RF, 99.97% accuracy for DFA in XGBoost and 99.94% accuracy for RWE in BootAgg. 4. By using DL, in linear features, gamma asymmetry has given more than 96% accuracy in RNN and 91% accuracy in LSTM and for non-linear features, 89% accuracy has been achieved for CD and AE in LSTM. 5. By combining linear and non-linear features, the highest accuracy was achieved in Bagging RF (98.50%) gamma asymmetry + RWE. In DL, Alpha + RWE, Gamma asymmetry + CD and gamma asymmetry + RWE have achieved 98% accuracy in LSTM.

A novel dataset was collected from the Central Institute of Psychiatry (CIP), Ranchi which was recorded using a 128-channels whereas major previous studies used fewer channels; the details of the study participants are summarized and a model is developed for statistical analysis using N-way ANOVA; artifacts are removed by high and low pass filtering of epoch data followed by re-referencing and independent component analysis for noise removal; linear features, namely, band power and interhemispheric asymmetry and non-linear features, namely, relative wavelet energy, wavelet entropy, Approximate entropy, sample entropy, detrended fluctuation analysis and correlation dimension are extracted; this model utilizes Epoch (213,072) for 5 s EEG data, which allows the model to train for longer, thereby increasing the efficiency of classifiers. Features scaling is done using a standard scalar rather than normalization because it helps increase the accuracy of the models (especially for deep learning algorithms) while PCA is used for feature reduction; the linear, non-linear and combination of both features are taken for extensive analysis in conjunction with ML and DL classifiers for the classification of depression. The combination of linear and non-linear features (only for those whose accuracy is highest) is used for the best detection results.

Despite the importance of cognitive monitoring, limited studies attempted to continuously monitor cognitive status of workers regarding mental fatigue effects on fall hazard. Thus, the objective of this study is to investigate and understand the effects of working at height on mental fatigue development for fall hazard prevention.

A quantitative framework using two well-known methods, i.e. Wavelet Packet Decomposition and Sample entropy, is developed to analyze the captured brain signals from Electroencephalography (EEG) to quantitatively assess mental fatigue levels, and seven mental fatigue indices were obtained. Between-subjects lab experiment was designed and conducted to assess mental fatigue in Virtual Reality (VR) environment.

Both of the quantitative methods confirmed that height exposure can adversely affect subjects' vigilance levels and indicated higher levels of mental fatigue. Significant differences were found between the two tested groups (i.e. working at height or on the ground) for six out of seven indices. The results suggested that working-at-height group had higher mental fatigue levels.

One limitation of this study is the limited number of subjects recruited for the experiment. Overall, this study is a preliminary and exploratory work towards mental fatigue monitoring and assessment in subjects exposed to fall risk.

This is the first study to explore and focus on mental fatigue assessment, particularly for construction falling-from-height hazard prevention by continuously monitoring mental fatigue levels of workers. The research provides insight into construction safety enhancement using smart technologies.

Exploiting an innovative measure of corporate culture based on machine learning and earnings conference calls, this study aims to investigate how corporate culture is influenced by hostile takeover threats. To sidestep endogeneity, this study uses a unique measure of takeover vulnerability principally based on the staggered implementation of state legislations, which are plausibly exogenous.

In addition to the standard regression analysis, this study also executes a variety of other empirical tests such as propensity score matching, entropy balancing and an instrumental variable analysis, to demonstrate that the results are robust. The final sample includes 27,663 firm-year observations from 4,092 distinct companies from 2001 to 2014.

This study documents that more takeover exposure weakens corporate culture considerably, consistent with the managerial myopia hypothesis. Threatened by the takeover risk, managers tend to behave myopically and are less likely to make long-term investments that promote strong corporate culture in the long run. Additional analysis focusing on a culture of innovation, which is especially vulnerable to managerial myopia, produces similar evidence.

To the best of the authors’ knowledge, this study is the first to explore the effect of takeover susceptibility on corporate culture using a distinctive metric of corporate culture based on textual analysis.

Exploiting novel measures of climate change exposure and corporate culture generated by a powerful textual analysis of earnings conference calls, this study aims to explore the effect of firm-specific climate change exposure on corporate innovation through the lens of corporate culture.

The authors apply the standard regression analysis as well as a variety of sophisticated techniques, namely, propensity score matching, entropy balancing and an instrumental-variable analysis with multiple alternative instruments.

The authors find that more exposure to climate change risk results in more innovation, as indicated by a significantly stronger culture of innovation. The findings are consistent with the notion that firms more exposed to climate change risk are pressed to be more innovative to adapt to the numerous changes caused by climate change. Finally, the authors also find that the effect of firm-level exposure on innovation is considerably less pronounced during uncertain times.

The authors are among the first studies to take advantage of a novel measure of firm-specific exposure to climate change and investigate how climate change exposure influences an innovative culture. Since climate change is a timely issue, the findings offer important implication to several stakeholders, such as shareholders, executives and investors in general.

Exploiting a novel measure of hostile takeover exposure primarily based on the staggered adoption of state legislations, we explore a crucial, albeit largely overlooked, aspect of corporate social responsibility (CSR). In particular, we investigate CSR inequality, which is the inequality across different CSR categories. Higher inequality suggests a less balanced, more lopsided, CSR policy.

In addition to the standard regression analysis, we perform several robustness checks including propensity score matching, entropy balancing and an instrumental-variable analysis.

Our results show that more takeover exposure exacerbates CSR inequality. Specifically, a rise in takeover vulnerability by one standard deviation results in an increase in CSR inequality by 4.53–5.40%. The findings support the managerial myopia hypothesis, where myopic managers promote some CSR activities that are useful to them in the short run more than others, leading to higher CSR inequality.

Our study is the first to exploit a unique measure of takeover vulnerability to investigate the impact of takeover threats on CSR inequality, which is an important aspect of CSR that is largely overlooked in the literature. We aptly fill this void in the literature.