Hierarchical semi-markov conditional random fields for recursive sequential data

Inspired by the hierarchical hidden Markov models (HHMM), we present the hierarchical semi-Markov conditional random field (HSCRF), a generalisation of embedded undirected Markov chains to model complex hierarchical, nested Markov processes. It is parameterised in a discriminative framework and has polynomial time algorithms for learning and inference. Importantly, we develop efficient algorithms for learning and constrained inference in a partially-supervised setting, which is important issue in practice where labels can only be obtained sparsely. We demonstrate the HSCRF in two applications: (i) recognising human activities of daily living (ADLs) from indoor surveillance cameras, and (ii) noun-phrase chunking. We show that the HSCRF is capable of learning rich hierarchical models with reasonable accuracy in both fully and partially observed data cases.

Mining sequential causal patterns with user-specified skeletons in multi-sequence of event data

The approaches proposed in the past for discovering sequential patterns mainly focused on single sequential data. In the real world, however, some sequential patterns hide their essences among multi-sequential event data. It has been noted that knowledge discovery with either user-specified constraints, or templates, or skeletons is receiving wide attention because it is more efficient and avoids the tedious selection of useful patterns from the mass-produced results. In this paper, a novel pattern in multi-sequential event data that are correlated and its mining approach are presented. We call this pattern sequential causal pattern. A group of skeletons of sequential causal patterns, which may be specified by the user or generated by the program, are verified or mined by embedding them into the mining engine. Experiments show that this method, when applied to discovering the occurring regularities of a crop pest in a region, is successful in mining sequential causal patterns with user-specified skeletons in multi-sequential event data.

Are OECD stock prices characterized by a random walk? Evidence from sequential trend break and panel data models

This paper examines whether stock prices for a sample of 22 OECD countries can be best represented as mean reversion or random walk processes. A sequential trend break test proposed by Zivot and Andrews is implemented, which has the advantage that it can take account of a structural break in the series, as well as panel data unit root tests proposed by Im et al., which exploits the extra power in the panel properties of the data. Results provide strong support for the random walk hypothesis.

Application of sequential nonparametric confidence bands in finance

In a nonparametric setting, the functional form of the relationship between the response variable and the associated predictor variables is assumed to be unknown when data is fitted to the model. Non-parametric regression models can be used for the same types of applications such as estimation, prediction, calibration, and optimization that traditional regression models are used for. The main aim of nonparametric regression is to highlight an important structure in the data without any assumptions about the shape of an underlying regression function. Hence the nonparametric approach allows the data to speak for itself. Applications of sequential procedures to a nonparametric regression model at a given point are considered.

The primary goal of sequential analysis is to achieve a given accuracy by using the smallest possible sample sizes. These sequential procedures allow an experimenter to make decisions based on the smallest number of observations without compromising accuracy. In the nonparametric regression model with a random design based on independent and identically distributed pairs of observations (X ,Y ), where the regression function m(x) is given bym(x) = E(Y X = x), estimation of the Nadaraya-Watson kernel estimator (m (x)) NW and local linear kernel estimator (m (x)) LL for the curve m(x) is considered. In order to obtain asymptotic confidence intervals form(x), two stage sequential procedure is used under which some asymptotic properties of Nadaraya-Watson and local linear estimators have been obtained.

The proposed methodology is first tested with the help of simulated data from linear and nonlinear functions. Encouraged by the preliminary findings from simulation results, the proposed method is applied to estimate the nonparametric regression curve of CAPM.

Sequential fixed-width confidence bands for kernel regression estimation

We consider a random design model based on independent and identically distributed (iid) pairs of observations (Xi, Yi), where the regression function m(x) is given by m(x) = E(Yi|Xi = x) with one independent variable. In a nonparametric setting the aim is to produce a reasonable approximation to the unknown function m(x) when we have no precise information about the form of the true density, f(x) of X. We describe an estimation procedure of non-parametric regression model at a given point by some appropriately constructed fixed-width (2d) confidence interval with the confidence coefficient of at least 1−. Here, d(> 0) and 2 (0, 1) are two preassigned values. Fixed-width confidence intervals are developed using both Nadaraya-Watson and local linear kernel estimators of nonparametric regression with data-driven bandwidths.

The sample size was optimized using the purely and two-stage sequential procedure together with asymptotic properties of the Nadaraya-Watson and local linear estimators. A large scale simulation study was performed to compare their coverage accuracy. The numerical results indicate that the confidence bands based on the local linear estimator have the best performance than those constructed by using Nadaraya-Watson estimator. However both estimators are shown to have asymptotically correct coverage properties.

Sample size determination for kernel regression estimation using sequential fixed-width confidence bands

We consider a random design model based on independent and identically distributed pairs of observations (Xi, Yi), where the regression function m(x) is given by m(x) = E(Yi|Xi = x) with one independent variable. In a nonparametric setting the aim is to produce a reasonable approximation to the unknown function m(x) when we have no precise information about the form of the true density, f(x) of X. We describe an estimation procedure of non-parametric regression model at a given point by some appropriately constructed fixed-width (2d) confidence interval with the confidence coefficient of at least 1−. Here, d(> 0) and 2 (0, 1) are two preassigned values. Fixed-width confidence intervals are developed using both Nadaraya-Watson and local linear kernel estimators of nonparametric regression with data-driven bandwidths. The sample size was optimized using the purely and two-stage sequential procedures together with asymptotic properties of the Nadaraya-Watson and local linear estimators. A large scale simulation study was performed to compare their coverage accuracy. The numerical results indicate that the confi dence bands based on the local linear estimator have the better performance than those constructed by using Nadaraya-Watson estimator. However both estimators are shown to have asymptotically correct coverage properties.

Part-time employment trends for full-time undergraduate property and construction students : results of a two year sequential investigation

This paper reports on the initial findings of an investigation into the experiences and part-time work practices of undergraduates enrolled in the Bachelor of Property and Construction (BPC) and combined degree courses at the University of Melbourne. Initial data was collected from final year students during 2001 and for all four years of the course in 2002. The results suggest that students in earlier years of the course are more likely to work in non-industry (casual) related employment and work fewer hours. Students in later years of the BPC course are more likely to work in jobs in the construction industry and also to work longer hours than those in earlier years. An analysis of final year students shows that the students employed by contracting organisations work significantly more hours than students employed by other types of organisations including architectural practices. The consequences of part-time semester employment on academic performance and students' well-being are considered and proposals are put forward as to how to better manage the industry experience-University relationship.

Obtaining accurate frequencies of sequential patterns over a single sequence

In the mining and analysis of a single long sequence, one fundamental and important problem is obtaining accurate frequencies of sequential patterns over the sequence. However, we identify that five previous frequency measures suffer from inherent inaccuracies. To obtain more accurate frequencies, we introduce two basic principles called strict anti-monotonicity and maximum-count for frequency measures. Under the two principles, a new frequency measure is presented. An algorithm is also devised to compute it. Both theoretical analysis and empirical evaluation show that more accurate frequencies can be obtained under the new measure

Acceleration data reveal the energy management strategy of a marine ectotherm during reproduction

1. Maintaining a high and stable body temperature is often critical for female ectotherms during reproduction. Yet this strategy may be energetically costly, and therefore challenging, during this period of already high-energy demand. 2. Here, the 6-week deployment of tri-axial accelerometers (n = 6) on a marine ectotherm, the loggerhead turtle (Caretta caretta), reproducing at the northern limit of the species’ breeding range (i.e. in a thermally dynamic environment) revealed the behavioural mechanisms underlying its energy management strategy during the breeding season. 3. The estimated activity levels of female loggerheads using overall dynamic body acceleration (ODBA) were high during the breeding season, suggesting that marine turtles may not be able to remain inactive for long periods in the same manner as terrestrial ectotherms, because of the thermally dynamic nature of their environment. 4. However, activity levels were not constant throughout the season, being impacted by both ambient water temperature and female reproductive status. In cold water at the beginning of the nesting season, high levels of activity suggested that females behaviourally thermoregulated by seeking out warm water patches along the shoreline. Interactions with male turtles (courtship and/or avoidance) may also explain this high level of activity. As sea temperatures warmed up and the amount of energy devoted to reproduction probably increased, the turtles spent more time resting during long sequential flat-bottomed dives, and reduced any unnecessary locomotory activity. 5. Turtles may therefore adjust their activity patterns in response to seasonal variations in abiotic (i.e. ambient temperature) and biotic (i.e. reproductive status) factors. This may help minimize activity-linked metabolic rate and maximize reproductive output over a season while breeding in thermally dynamic environments. 6. A mechanistic model gave support to these empirical results. The model revealed that actively maintaining high and stable body temperature is of clear benefit to female turtles at temperate breeding sites. While energetically costly, such active thermoregulatory behaviour may speed up egg maturation, allowing turtles to initiate nesting earlier in the season, and hence maximize reproductive output.

Recommendations for improved data processing from expired gas analysis indirect calorimetry

There is currently no universally recommended and accepted method of data processing within the science of indirect calorimetry for either mixing chamber or breath-by-breath systems of expired gas analysis. Exercise physiologists were first surveyed to determine methods used to process oxygen consumption ([OV0312]O 2) data, and current attitudes to data processing within the science of indirect calorimetry. Breath-by-breath datasets obtained from indirect calorimetry during incremental exercise were then used to demonstrate the consequences of commonly used time, breath and digital filter post-acquisition data processing strategies. Assessment of the variability in breath-by-breath data was determined using multiple regression based on the independent variables ventilation (VE), and the expired gas fractions for oxygen and carbon dioxide, FEO 2 and FECO2, respectively. Based on the results of explanation of variance of the breath-by-breath [OV0312]O2 data, methods of processing to remove variability were proposed for time-averaged, breath-averaged and digital filter applications. Among exercise physiologists, the strategy used to remove the variability in sequential [OV0312]O2 measurements varied widely, and consisted of time averages (30 sec [38%], 60 sec [18%], 20 sec [11%], 15 sec [8%]), a moving average of five to 11 breaths (10%), and the middle five of seven breaths (7%). Most respondents indicated that they used multiple criteria to establish maximum [OV0312]O 2 ([OV0312]O2max) including: the attainment of age-predicted maximum heart rate (HRmax) [53%], respiratory exchange ratio (RER) >1.10 (49%) or RER >1.15 (27%) and a rating of perceived exertion (RPE) of >17, 18 or 19 (20%). The reasons stated for these strategies included their own beliefs (32%), what they were taught (26%), what they read in research articles (22%), tradition (13%) and the influence of their colleagues (7%). The combination of VE, FEO 2 and FECO2 removed 96-98% of [OV0312]O2 breath-by-breath variability in incremental and steady-state exercise [OV0312]O2 data sets, respectively. Correction of residual error in [OV0312]O2 datasets to 10% of the raw variability results from application of a 30-second time average, 15-breath running average, or a 0.04 Hz low cut-off digital filter. Thus, we recommend that once these data processing strategies are used, the peak or maximal value becomes the highest processed datapoint. Exercise physiologists need to agree on, and continually refine through empirical research, a consistent process for analysing data from indirect calorimetry.

A sequential purchasing power parity test for panels of large cross-sections and implications for investors

In this paper we use monthly time series data for not less than 64 countries and a new sequential approach to test for purchasing power parity (PPP). The results are strong in that the evidence in favor of PPP is very weak. In fact, for the US-dollar-based exchange rates the evidence is basically non-existent. In order to eliminate the effect of the base currency, we also apply the sequential PPP test to all pairs of exchange rates, and find similarly weak evidence of PPP. However, for those rates where evidence is found, using a technical trading rule, we find evidence of significant profits. The predictability of the stationary pairs is therefore important for investors. © 2014 © 2014 Taylor & Francis.

Exposing HPC and sequential applications as services through the development and deployment of a SaaS cloud

Cloud and service computing has started to change the way research in science, in particular biology and medicine, is being carried out. Researchers that have taken advantage of this technology (making use of public and private cloud compute resources) can process large amounts of data (big data) and speed up discovery. However, this requires researchers to acquire a solid knowledge and skills in the development of sequential and high performance computing (HPC), and cloud development and deployment background. In response a technology exposing HPC applications as services through the development and deployment of a SaaS cloud, and its proof of concept in the form of implementation of a cloud environment, Uncinus, has been developed and implemented to allow researchers easy access to cloud computing resources. The new technology offers and Uncinus supports the development of applications as services and the sharing of compute resources to speed up applications' execution. Users access these cloud resources and services through web interfaces. Using the Uncinus platform, a bio-informatics workflow was executed on a private (HPC) cloud, server and public cloud (Amazon EC2) resources, performance results showing a 3 fold improvement compared to local resources' performance. Biology and medicine specialists with no programming and application deployment on clouds background could run the case study applications with ease.

A sequential purchasing power parity test for panels of large cross-sections and implications for investors

In this paper we use monthly time series data for not less than 64 countries and a new sequential approach to test for purchasing power parity (PPP). The results are strong in that the evidence in favor of PPP is very weak. In fact, for the US-dollar-based exchange rates the evidence is basically non-existent. In order to eliminate the effect of the base currency, we also apply the sequential PPP test to all pairs of exchange rates, and find similarly weak evidence of PPP. However, for those rates where evidence is found, using a technical trading rule, we find evidence of significant profits. The predictability of the stationary pairs is therefore important for investors.

Sequential polyandry through divorce and re-pairing in a cooperatively breeding bird reduces helper-offspring relatedness

Polyandry is an important component of both sexual selection and kin structuring within cooperatively breeding species. A female may have multiple partners within a single reproductive attempt (simultaneous polyandry) or across multiple broods within and/or across years (sequential polyandry). Both types of polyandry confer a range of costs and benefits to individuals and alter the genetic structure of social groups over time. To date, many molecular studies of cooperative breeders have examined the evolution of cooperative breeding in relation to simultaneous polyandry. However, cooperatively breeding vertebrates are iteroparous, and thus sequential polyandry is also likely, but more rarely considered in this context. We examined sequential polyandry in a cooperatively breeding bird that has a low level of within-brood polyandry. Over a 5-year period (2006–2010), we monitored individual mating relationships using molecular markers in a population of individually marked apostlebirds (Struthidea cinerea). Divorce occurred between reproductive seasons in 17 % (8/48) of pairs and appeared to be female-driven. The level of sequential polyandry was also driven by the disappearance of males after breeding, and over 90 % of females, for whom we had suitable data, bred with multiple males over the period of study. This sequential polyandry significantly altered the relatedness of group members to the offspring in the nest. However, in about half of the cases, the second male was related (first- or second-order relative) to the first male of a sequentially polyandrous female and this alleviated the reduction in relatedness caused by polyandry. Our findings suggest that even in species with high within-brood parentage certainty, helper-offspring relatedness values can quickly erode through sequential polyandry.

Managing undesirable outputs in the Australian construction industry using Data Envelopment Analysis models

Carbon reduction has become one of the most significant challenges for economic growth. This paper presents the preliminary analysis of undesirable output reduction targets and emission schedules in temporal-spatial comparisons based on Data Envelopment Analysis. The reduction targets of undesirable outputs are investigated, including the maximum, input, technical and ideal reduction targets. Four Data Envelopment Analysis models that are based on a sequential benchmark technology and variable returns to scale are introduced to measure these reduction percentages. In order to formulate the optimal emission schedule of undesirable outputs, an optimal model is provided without inflation. Data from the Australian construction industry from 2000 to 2010 are employed to develop the models. The results of the analysis indicate that the Australian Government could achieve its promised carbon reduction targets in the construction industry. Most Australian regions' construction industries possess carbon mitigation potential and some of them could increase their desirable outputs if carbon were decreased. This paper suggests that policymakers can benefit from formulating various suitable undesirable output reduction objectives and schedules through the models developed. The research method can be replicated for other sectors and regions focussing on undesirable output reduction.