A hierarchical PCA-based anomaly detection model

A hierarchical intrusion detection model is proposed to detect both anomaly and misuse attacks. In order to further speed up the training and testing, PCA-based feature extraction algorithm is used to reduce the dimensionality of the data. A PCA-based algorithm is used to filter normal data out in the upper level. The experiment results show that PCA can reduce noise in the original data set and the PCA-based algorithm can reach the desirable performance.

Consumer response to unit pricing for mobile telephony

This project is a co-operative study between ACCAN and Deakin University. It focuses on Unit Pricing, the practice of displaying the price of goods or services based on a standard quantity, to allow a direct comparison between competitive offers. This study aimed at gauging whether the new unit pricing information for mobile phone contracts assists consumers in assessing and comparing the value provided across alternative contracts within and between suppliers. Some 24 in-depth interviews were conducted with consumers who had recently bought or renewed a mobile phone contract.
The research showed that most consumers could use unit pricing information and some found it useful. Where consumers’ plans had unlimited or infinite capacity, unit pricing information was not relevant. Many consumers preferred voice allowances to be expressed in minutes, rather than in dollar allowances. Data was the most problematic category, as consumers typically had only limited understanding of the amount of data that various applications used. Most did have a broad understanding of what total capacity in data they would need, typically expressed in gigabytes.
Consumers commonly sought simplicity in deciding on which plan they would purchase or renew. A key issue for consumers was not “going over”, that is not exceeding their call, text or data allowances. For that reason, they were prepared to choose a plan that commonly resulted in them not using their full allowances each month. Some consumers used Apps on their smartphones to monitor their usage. Not all consumers had experienced advisory messages about nearing the limits of their plan’s allowances.
The Report recommended that:

R1. Unit pricing should be maintained
R2. Where unit pricing is provided for call costs, these should be expressed in terms of a one-minute call.
R3 Unit pricing for data should be expressed in terms of gigabytes or part thereof.
R4 In advertising mobile phone plans and at point of sales, customers should be provided with three levels of information – 1) overall plan features, 2) unit pricing information and 3) a data calculator.
R5 Level 2 and 3 information should be provided in a standard format across the industry, enabling consumers to make ready comparisons between plans and between competitive offers from different providers.
R6. Continuing public education is needed.
R7. Warnings about going over should always include the date when the allowance period ends and tell consumers what the rate will be if they “go over” based on the Level 2 information.
R8. The Consumer Protection Code should be reviewed in the light of these findings and recommendations.

The pricing of deposit insurance in the presence of systematic risk

Based on the Merton (1977) put option framework, we develop a deposit insurance pricing model that incorporates asset correlations, a measurement for the systematic risk of a bank, to account for the risk of joint bank failures. Estimates from our model suggest that actuarially fair risk-based deposit insurance that considers only individual bank failure risk is underpriced, leaving insurance providers exposed to net losses. Our estimates also capture the size premium where big banks are priced with higher deposit insurance than small banks. This result is particularly relevant to the current regulatory concerns on big banks that are too-big-to-fail. Above all, our approach provides a unifying framework for integrating risk-based deposit insurance with risk-based Basel capital requirements.

Supporting working parents: findings from the AHRC’s national inquiry into the prevalence of pregnancy and return to work discrimination in the workplace

In July 2014 the Australian Human Rights Commission (AHRC) released the findings of its national review into pregnancy and return to work discrimination in the workplace1 which it conducted following a request from the Commonwealth Attorney-General’s Department.2 The review comes 15 years after the commission’s first inquiry into pregnancy discrimination in the workplace.3Federal law has prohibited pregnancy discrimination in the workplace since the Sex Discrimination Act 1984 (Cth) (SDA) came into force.4 It is now unlawful in every state and territory.5 Discrimination on the basis of breastfeeding and family or carer’s responsibilities is also prohibited.6 Since 2009 the Fair Work Act 2009 (Cth) (FW Act) has prohibited workplace discrimination based on pregnancy and family or carer’s responsibilities7 and the Act gives employees additional entitlements relating to their parental and caring responsibilities. Male and female employees who are the primary caregiver for a child are entitled to 12 months unpaid parental leave upon the birth or adoption of the child and can request an additional 12 months leave.8 Upon returning to work, they can request flexible working conditions9 and they are protected from adverse action, such as dismissal, for exercising these rights.10 Yet despite these legal protections, the findings of the national review show that employees continue to experience discrimination during pregnancy, when taking parental leave and upon re-entering the workforce. This note presents the main findings from the surveys and consultations that were held with employers and employees as part of the review and the review’s recommendations for addressing the prevalence of what it terms ‘pregnancy/return to work discrimination’.

Use of motivational interviewing to improve return-to-work and work-related outcomes: a review

Work-based return-to-work (RTW) interventions can help to reduce the duration and cost of work disability, and in turn, prevent the negative effects of long-term sickness absence. However, there are a number of complex cognitive, affective and behavioural factors that can impact an individual's confidence, motivation and willingness to RTW that need to be addressed to facilitate effective outcomes. This literature review investigates evidence for the use of motivational interviewing (MI) for improving return-to-work (RTW) and employment outcomes. Whilst evidence for the efficacy of MI in clinical settings to motivate health behaviour change is strong, more research is needed to determine whether MI can be usefully applied to improve RTW and other work-related outcomes. © 2014 The Author(s).

Anisotropic hardening model based on non-associated flow rule and combined nonlinear kinematic hardening for sheet materials

 A material model for more effective analysis of plastic deformation of sheet materials is presented in this paper. The model is capable of considering the following aspects of plastic deformation behavior of sheet materials: the anisotropy in yielding stresses in different directions by using a quadratic yield function (based on Hill’s 1948 model and stress ratios), the anisotropy in work hardening by introducing non-constant flow stress hardening in different directions, the anisotropy in plastic strains in different directions by using a quadratic plastic potential function and non-associated flow rule (based on Hill’s 1948 model and plastic strain ratios, r-values), and finally some of the cyclic hardening phenomena such as Bauschinger’s effect and transient behavior for reverse loading by using a coupled nonlinear kinematic hardening (so-called Armstrong-Frederick-Chaboche model). Basic fundamentals of the plasticity of the model are presented in a general framework. Then, the model adjustment procedure is derived for the plasticity formulations. Also, a generic numerical stress integration procedure is developed based on backward-Euler method (so-called multistage return mapping algorithm). Different aspects of the model are verified for DP600 steel sheet. Results show that the new model is able to predict the sheet material behavior in both anisotropic hardening and cyclic hardening regimes more accurately. By featuring the above-mentioned facts in the presented constitutive model, it is expected that more accurate results can be obtained by implementing this model in computational simulations of sheet material forming processes. For instance, more precise results of springback prediction of the parts formed from highly anisotropic hardened materials or that of determining the forming limit diagrams is highly expected by using the developed material model.

Estimating Forward Pricing Function: How Efficient is Indian Stock Index Futures Market?

This paper uses Indian stock futures data to explore unbiased expectations and efficient market hypothesis. Having experienced voluminous transactions within a short time span after its establishment, the Indian stock futures market provides an unparalleled case for exploring these issues involving expectation and efficiency. Besides analyzing market efficiency between cash and futures prices using cointegration and error correction frameworks, the efficiency hypothesis is also investigated after explicitly modeling the underlying state of the market (expansion or contraction) through the first-order Markov switching set-up. The results based on Markov switching analysis show that relatively longer time horizon is more effective in eliminating arbitrage opportunities than the short run.

City-level auditor industry specialization, economies of scale, and audit pricing

We examine the effects of city-level auditor industry specialization and scale economies on audit pricing in the United States. Using a sample of Big N clients for the 2000-2007 period, and a scale measure based on percentile rankings of the number of audit clients at the city-industry level, we document significant specialization premiums and scale discounts in both the pre- and post-Sarbanes-Oxley Act (SOX) periods. However, the effects of industry specialization and scale economies on audit pricing are highly interactive. The negative effect of city-industry scale on audit fees obtains only for clients of specialist auditors. By contrast, clients of non-specialist auditors obtain scale discounts only when they enjoy strong bargaining power, suggesting that auditors are "forced" to pass on scale economies to clients with greater bargaining power.

Executive equity risk-taking incentives and audit pricing

Using a large sample of U.S. firms spanning the period 2000-2010, we document a strong positive association between the sensitivity of CEO compensation portfolio to stock return volatility (vega) and audit fees. We also show that the positive association between vega and audit fees is weaker in the post-Sarbanes-Oxley Act (SOX) period. In supplementary tests, we show that the relation between vega and audit fees is stronger for firms with older CEOs and in firms where the CEO is also chairman of the board. Collectively, our results suggest that audit firms incorporate executive risktaking incentives in the fees they charge for their services.

Derivation of dynamic material properties of steel-fibre-reinforced concrete using kernel regression

The reliable and efficient design of steel-fibre-reinforced concrete (SFRC) structures requires clear knowledge of material properties. Since the locations and orientations of aggregates and fibres in concrete are intrinsically random, testing results from different specimens vary, and it needs hundreds or even thousands of specimens and tests to derive the unbiased statistical distributions of material properties by using traditional statistical techniques. Therefore, few statistical studies on the SFRC material properties can be found in literature. In this study, high-rate impact test results on SFRC using split Hopkinson pressure bar are further analysed. The influences of different strain rates and various volume fractions of fibres on compressive strength of SFRC specimens under dynamic loadings will be quantified, by using kernel regression, a kernel-based nonparametric statistical method. Several kernel estimators and functions will be compared. This technique allows one to derive an unbiased statistical estimation from limited testing data. Therefore it is especially useful when the testing data is limited.

Stock return predictability and determinants of predictability and profits

We examine stock return predictability for India and find strong evidence of sectoral return predictability over market return predictability. We show that mean-variance investors make statistically significant and economically meaningful profits by tracking financial ratios. For the first time in this literature, we examine the determinants of time-varying predictability and mean-variance profits. We show that both expected and unexpected shocks emanating from most financial ratios explain sectoral return predictability and profits. These are fresh contributions to the understanding of asset pricing.

Auditee religiosity, external monitoring, and the pricing of audit services

Based on prior studies which show that firms headquartered in high religiosity counties exhibit high level of business ethics, this study examines whether these firms are associated with low audit risk, and therefore low audit fees. In investigating this relationship, we draw a distinction between intrinsic and extrinsic religiosity of auditees. Using a sample of 25,872 U.S. observations from 2003 to 2012, we find that intrinsic religiosity of the auditees is associated with low audit fees after controlling for auditee extrinsic religiosity, social capital, firm-specific characteristics, and county-specific characteristics. Furthermore, we find that external monitoring (institutional ownership and leverage) weakens the negative relationship between auditee intrinsic religiosity and audit fees. Finally, we conclude that the effect of auditor religiosity on audit fees is a regional effect that may affect the relationship between audit fees and auditee intrinsic religiosity.

A comparative analysis of accounting-based valuation models

We empirically compare the reliability of the dividend (DIV) model, the residual income valuation (CT, GLS) model, and the abnormal earnings growth (OJ) model. We find that valuation estimates from the OJ model are generally more reliable than those from the other three models, because the residual income valuation model anchored by book value gets off to a poor start when compared with the OJ model led by capitalized next-year earnings. We adopt a 34-year sample covering from 1985 to 2013 to compare the reliability of valuation estimates via their means of absolute pricing errors (MAPE) and corresponding t statistics. We further use the switching regression of Barrios and Blanco to show that the average probability of OJ valuation estimates is greater in explaining stock prices than the DIV, CT, and GLS models. In addition, our finding that the OJ model yields more reliable estimates is robust to analysts-based and model-based earnings measures.

Robustness and embedding capacity enhancement in time-spread echo-based audio watermarking

In echo-based audio watermarking methods, poor robustness and low embedding capacity are the main problems. In this paper, we propose a novel time-spread echo method for audio watermarking, aiming to improve the robustness and the embedding capacity. To improve the robustness, we design an efficient pseudonoise (PN) sequence and a corresponding decoding function. Compared to the conventional PN sequence used in time-spread echo hiding based method, more large peaks are produced during the autocorrelation of the proposed PN sequence. Our decoding function is designed to utilize these peaks to improve the robustness. To enhance the embedding capacity, multiple watermark bits are embedded into one audio segment. This is achieved by varying the delays of added echo signals. Moreover, the security of the proposed method is further improved by scrambling the watermarks at the embedding stage. Compared with the conventional time-spread echo-based method, the proposed method is more robust to conventional attacks and has higher embedding capacity. The effectiveness of our method is illustrated by simulation results.