Sparse model construction using coordinate descent optimization

We propose a new sparse model construction method aimed at maximizing a model’s generalisation capability for a large class of linear-in-the-parameters models. The coordinate descent optimization algorithm is employed with a modified l1- penalized least squares cost function in order to estimate a single parameter and its regularization parameter simultaneously based on the leave one out mean square error (LOOMSE). Our original contribution is to derive a closed form of optimal LOOMSE regularization parameter for a single term model, for which we show that the LOOMSE can be analytically computed without actually splitting the data set leading to a very simple parameter estimation method. We then integrate the new results within the coordinate descent optimization algorithm to update model parameters one at the time for linear-in-the-parameters models. Consequently a fully automated procedure is achieved without resort to any other validation data set for iterative model evaluation. Illustrative examples are included to demonstrate the effectiveness of the new approaches.

Improving the evaluation of hydrological multi-model forecast performance in the Upper Danube Catchment

Medium range flood forecasting activities, driven by various meteorological forecasts ranging from high resolution deterministic forecasts to low spatial resolution ensemble prediction systems, share a major challenge in the appropriateness and design of performance measures. In this paper possible limitations of some traditional hydrological and meteorological prediction quality and verification measures are identified. Some simple modifications are applied in order to circumvent the problem of the autocorrelation dominating river discharge time-series and in order to create a benchmark model enabling the decision makers to evaluate the forecast quality and the model quality. Although the performance period is quite short the advantage of a simple cost-loss function as a measure of forecast quality can be demonstrated.

Assessing the impact on chronic disease of incorporating the societal cost of greenhouse gases into the price of food: an econometric and comparative risk assessment modelling study

Objectives To model the impact on chronic disease of a tax on UK food and drink that internalises the wider costs to society of greenhouse gas (GHG) emissions and to estimate the potential revenue. Design An econometric and comparative risk assessment modelling study. Setting The UK. Participants The UK adult population. Interventions Two tax scenarios are modelled: (A) a tax of £2.72/tonne carbon dioxide equivalents (tCO2e)/100 g product applied to all food and drink groups with above average GHG emissions. (B) As with scenario (A) but food groups with emissions below average are subsidised to create a tax neutral scenario. Outcome measures Primary outcomes are change in UK population mortality from chronic diseases following the implementation of each taxation strategy, the change in the UK GHG emissions and the predicted revenue. Secondary outcomes are the changes to the micronutrient composition of the UK diet. Results Scenario (A) results in 7770 (95% credible intervals 7150 to 8390) deaths averted and a reduction in GHG emissions of 18 683 (14 665to 22 889) ktCO2e/year. Estimated annual revenue is £2.02 (£1.98 to £2.06) billion. Scenario (B) results in 2685 (1966 to 3402) extra deaths and a reduction in GHG emissions of 15 228 (11 245to 19 492) ktCO2e/year. Conclusions Incorporating the societal cost of GHG into the price of foods could save 7770 lives in the UK each year, reduce food-related GHG emissions and generate substantial tax revenue. The revenue neutral scenario (B) demonstrates that sustainability and health goals are not always aligned. Future work should focus on investigating the health impact by population subgroup and on designing fiscal strategies to promote both sustainable and healthy diets.

Cost effectiveness of mitigation measures for water security and water quality in the River Thames catchment: impacts of changing climate, land use and water scarcity

The catchment of the River Thames, the principal river system in southern England, provides the main water supply for London but is highly vulnerable to changes in climate, land use and population. The river is eutrophic with significant algal blooms with phosphorus assumed to be the primary chemical indicator of ecosystem health. In the Thames Basin, phosphorus is available from point sources such as wastewater treatment plants and from diffuse sources such as agriculture. In order to predict vulnerability to future change, the integrated catchments model for phosphorus (INCA-P) has been applied to the river basin and used to assess the cost-effectiveness of a range of mitigation and adaptation strategies. It is shown that scenarios of future climate and land-use change will exacerbate the water quality problems, but a range of mitigation measures can improve the situation. A cost-effectiveness study has been undertaken to compare the economic benefits of each mitigation measure and to assess the phosphorus reductions achieved. The most effective strategy is to reduce fertilizer use by 20% together with the treatment of effluent to a high standard. Such measures will reduce the instream phosphorus concentrations to close to the EU Water Framework Directive target for the Thames.

Fast identification algorithms for Gaussian process model

A class identification algorithms is introduced for Gaussian process(GP)models.The fundamental approach is to propose a new kernel function which leads to a covariance matrix with low rank,a property that is consequently exploited for computational efficiency for both model parameter estimation and model predictions.The objective of either maximizing the marginal likelihood or the Kullback–Leibler (K–L) divergence between the estimated output probability density function(pdf)and the true pdf has been used as respective cost functions.For each cost function,an efficient coordinate descent algorithm is proposed to estimate the kernel parameters using a one dimensional derivative free search, and noise variance using a fast gradient descent algorithm. Numerical examples are included to demonstrate the effectiveness of the new identification approaches.

The extent to which potential benefits to EU farmers of adopting transgenic crops are reduced by cost of compliance with coexistence regulations

This article forecasts the extent to which the potential benefits of adopting transgenic crops may be reduced by costs of compliance with coexistence regulations applicable in various member states of the EU. A dynamic economic model is described and used to calculate the potential yield and gross margin of a set of crops grown in a selection of typical rotation scenarios. The model simulates varying levels of pest, weed, and drought pressures, with associated management strategies regarding pesticide and herbicide application, and irrigation. We report on the initial use of the model to calculate the net reduction in gross margin attributable to coexistence costs for insect-resistant (IR) and herbicide-tolerant (HT) maize grown continuously or in a rotation, HT soya grown in a rotation, HT oilseed rape grown in a rotation, and HT sugarbeet grown in a rotation. Conclusions are drawn about conditions favoring inclusion of a transgenic crop in a crop rotation, having regard to farmers’ attitude toward risk.

An interdisciplinary modelling approach asessing the cost-effectiveness of agri-environmental measures on reducing nutrient concentration to WFD thresholds under climate change: the case of the Louros catchment

The application of the Water Framework Directive (WFD) in the European Union (EU) targets certain threshold levels for the concentration of various nutrients, nitrogen and phosphorous being the most important. In the EU, agri-environmental measures constitute a significant component of Pillar 2—Rural Development Policies in both financial and regulatory terms. Environmental measures also are linked to Pillar 1 payments through cross-compliance and the greening proposals. This paper drawing from work carried out in the REFRESH FP7 project aims to show how an INtegrated CAtchment model of plant/soil system dynamics and instream biogeochemical and hydrological dynamics can be used to assess the cost-effectiveness of agri-environmental measures in relation to nutrient concentration targets set by the WFD, especially in the presence of important habitats. We present the procedures (methodological steps, challenges and problems) for assessing the cost-effectiveness of agri-environmental measures at the baseline situation, and climate and land use change scenarios. Furthermore, we present results of an application of this methodology to the Louros watershed in Greece and discuss the likely uses and future extensions of the modelling approach. Finally, we attempt to reveal the importance of this methodology for designing and incorporating alternative environmental practices in Pillar 1 and 2 measures.

Aircraft routing with minimal climate impact: the REACT4C climate cost function modelling approach (V1.0)

In addition to CO2, the climate impact of aviation is strongly influenced by non-CO2 emissions, such as nitrogen oxides, influencing ozone and methane, and water vapour, which can lead to the formation of persistent contrails in ice-supersaturated regions. Because these non-CO2 emission effects are characterised by a short lifetime, their climate impact largely depends on emission location and time; that is to say, emissions in certain locations (or times) can lead to a greater climate impact (even on the global average) than the same emission in other locations (or times). Avoiding these climate-sensitive regions might thus be beneficial to climate. Here, we describe a modelling chain for investigating this climate impact mitigation option. This modelling chain forms a multi-step modelling approach, starting with the simulation of the fate of emissions released at a certain location and time (time-region grid points). This is performed with the chemistry–climate model EMAC, extended via the two submodels AIRTRAC (V1.0) and CONTRAIL (V1.0), which describe the contribution of emissions to the composition of the atmosphere and to contrail formation, respectively. The impact of emissions from the large number of time-region grid points is efficiently calculated by applying a Lagrangian scheme. EMAC also includes the calculation of radiative impacts, which are, in a second step, the input to climate metric formulas describing the global climate impact of the emission at each time-region grid point. The result of the modelling chain comprises a four-dimensional data set in space and time, which we call climate cost functions and which describes the global climate impact of an emission at each grid point and each point in time. In a third step, these climate cost functions are used in an air traffic simulator (SAAM) coupled to an emission tool (AEM) to optimise aircraft trajectories for the North Atlantic region. Here, we describe the details of this new modelling approach and show some example results. A number of sensitivity analyses are performed to motivate the settings of individual parameters. A stepwise sanity check of the results of the modelling chain is undertaken to demonstrate the plausibility of the climate cost functions.

MetUM-GOML: a near-globally coupled atmosphere–ocean-mixed-layer model

Well-resolved air–sea interactions are simulated in a new ocean mixed-layer, coupled configuration of the Met Office Unified Model (MetUM-GOML), comprising the MetUM coupled to the Multi-Column K Profile Parameterization ocean (MC-KPP). This is the first globally coupled system which provides a vertically resolved, high near-surface resolution ocean at comparable computational cost to running in atmosphere-only mode. As well as being computationally inexpensive, this modelling framework is adaptable– the independent MC-KPP columns can be applied selectively in space and time – and controllable – by using temperature and salinity corrections the model can be constrained to any ocean state. The framework provides a powerful research tool for process-based studies of the impact of air–sea interactions in the global climate system. MetUM simulations have been performed which separate the impact of introducing inter- annual variability in sea surface temperatures (SSTs) from the impact of having atmosphere–ocean feedbacks. The representation of key aspects of tropical and extratropical variability are used to assess the performance of these simulations. Coupling the MetUM to MC-KPP is shown, for example, to reduce tropical precipitation biases, improve the propagation of, and spectral power associated with, the Madden–Julian Oscillation and produce closer-to-observed patterns of springtime blocking activity over the Euro-Atlantic region.

Comparing hybrid data assimilation methods on the Lorenz 1963 model with increasing nonlinearity

We systematically compare the performance of ETKF-4DVAR, 4DVAR-BEN and 4DENVAR with respect to two traditional methods (4DVAR and ETKF) and an ensemble transform Kalman smoother (ETKS) on the Lorenz 1963 model. We specifically investigated this performance with increasing nonlinearity and using a quasi-static variational assimilation algorithm as a comparison. Using the analysis root mean square error (RMSE) as a metric, these methods have been compared considering (1) assimilation window length and observation interval size and (2) ensemble size to investigate the influence of hybrid background error covariance matrices and nonlinearity on the performance of the methods. For short assimilation windows with close to linear dynamics, it has been shown that all hybrid methods show an improvement in RMSE compared to the traditional methods. For long assimilation window lengths in which nonlinear dynamics are substantial, the variational framework can have diffculties fnding the global minimum of the cost function, so we explore a quasi-static variational assimilation (QSVA) framework. Of the hybrid methods, it is seen that under certain parameters, hybrid methods which do not use a climatological background error covariance do not need QSVA to perform accurately. Generally, results show that the ETKS and hybrid methods that do not use a climatological background error covariance matrix with QSVA outperform all other methods due to the full flow dependency of the background error covariance matrix which also allows for the most nonlinearity.

Obsessive Compulsive Treatment Efficacy Trial (OCTET) comparing the clinical and cost effectiveness of self-managed therapies: study protocol for a randomised controlled trial

Background: UK National Institute of Health and Clinical Excellence guidelines for obsessive compulsive disorder (OCD) specify recommendations for the treatment and management of OCD using a stepped care approach. Steps three to six of this model recommend treatment options for people with OCD that range from low-intensity guided self-help (GSH) to more intensive psychological and pharmacological interventions. Cognitive behavioural therapy (CBT), including exposure and response prevention, is the recommended psychological treatment. However, whilst there is some preliminary evidence that self-managed therapy packages for OCD can be effective, a more robust evidence base of their clinical and cost effectiveness and acceptability is required. Methods/Design: Our proposed study will test two different self-help treatments for OCD: 1) computerised CBT (cCBT) using OCFighter, an internet-delivered OCD treatment package; and 2) GSH using a book. Both treatments will be accompanied by email or telephone support from a mental health professional. We will evaluate the effectiveness, cost and patient and health professional acceptability of the treatments. Discussion: This study will provide more robust evidence of efficacy, cost effectiveness and acceptability of self-help treatments for OCD. If cCBT and/or GSH prove effective, it will provide additional, more accessible treatment options for people with OCD.

Do laws influence the cost of real estate brokerage services? A state fixed effects approach

A FTC-DOJ study argues that state laws and regulations may inhibit the unbundling of real estate brokerage services in response to new technology. Our data show that 18 states have changed laws in ways that promote unbundling since 2000. We model brokerage costs as measured by number of agents in a state-level annual panel vector autoregressive framework, a novel way of analyzing wasteful competition. Our findings support a positive relationship between brokerage costs and lagged house price and transactions. We find that change in full-service brokers responds negatively (by well over two percentage points per year) to legal changes facilitating unbundling

Decadal predictability of regional-scale peak winds over Europe using the Earth System Model of the Max-Planck-Institute for Meteorology

The predictability of high impact weather events on multiple time scales is a crucial issue both in scientific and socio-economic terms. In this study, a statistical-dynamical downscaling (SDD) approach is applied to an ensemble of decadal hindcasts obtained with the Max-Planck-Institute Earth System Model (MPI-ESM) to estimate the decadal predictability of peak wind speeds (as a proxy for gusts) over Europe. Yearly initialized decadal ensemble simulations with ten members are investigated for the period 1979–2005. The SDD approach is trained with COSMO-CLM regional climate model simulations and ERA-Interim reanalysis data and applied to the MPI-ESM hindcasts. The simulations for the period 1990–1993, which was characterized by several windstorm clusters, are analyzed in detail. The anomalies of the 95 % peak wind quantile of the MPI-ESM hindcasts are in line with the positive anomalies in reanalysis data for this period. To evaluate both the skill of the decadal predictability system and the added value of the downscaling approach, quantile verification skill scores are calculated for both the MPI-ESM large-scale wind speeds and the SDD simulated regional peak winds. Skill scores are predominantly positive for the decadal predictability system, with the highest values for short lead times and for (peak) wind speeds equal or above the 75 % quantile. This provides evidence that the analyzed hindcasts and the downscaling technique are suitable for estimating wind and peak wind speeds over Central Europe on decadal time scales. The skill scores for SDD simulated peak winds are slightly lower than those for large-scale wind speeds. This behavior can be largely attributed to the fact that peak winds are a proxy for gusts, and thus have a higher variability than wind speeds. The introduced cost-efficient downscaling technique has the advantage of estimating not only wind speeds but also estimates peak winds (a proxy for gusts) and can be easily applied to large ensemble datasets like operational decadal prediction systems.

Simulating the impact on health of internalising the cost of carbon in food prices combined with a tax on sugar-sweetened beverages

Rising greenhouse gas emissions (GHGEs) have implications for health and up to 30 % of emissions globally are thought to arise from agriculture. Synergies exist between diets low in GHGEs and health however some foods have the opposite relationship, such as sugar production being a relatively low source of GHGEs. In order to address this and to further characterise a healthy sustainable diet, we model the effect on UK non-communicable disease mortality and GHGEs of internalising the social cost of carbon into the price of food alongside a 20 % tax on sugar sweetened beverages (SSBs). Developing previously published work, we simulate four tax scenarios: (A) a GHGEs tax of £2.86/tonne of CO2 equivalents (tCO2e)/100 g product on all products with emissions greater than the mean across all food groups (0.36 kgCO2e/100 g); (B) scenario A but with subsidies on foods with emissions lower than 0.36 kgCO2e/100 g such that the effect is revenue neutral; (C) scenario A but with a 20 % sales tax on SSBs; (D) scenario B but with a 20 % sales tax on SSBs. An almost ideal demand system is used to estimate price elasticities and a comparative risk assessment model is used to estimate changes to non-communicable disease mortality. We estimate that scenario A would lead to 300 deaths delayed or averted, 18,900 ktCO2e fewer GHGEs, and £3.0 billion tax revenue; scenario B, 90 deaths delayed or averted and 17,100 ktCO2e fewer GHGEs; scenario C, 1,200 deaths delayed or averted, 18,500 ktCO2e fewer GHGEs, and £3.4 billion revenue; and scenario D, 2,000 deaths delayed or averted and 16,500 ktCO2e fewer GHGEs. Deaths averted are mainly due to increased fibre and reduced fat consumption; a SSB tax reduces SSB and sugar consumption. Incorporating the social cost of carbon into the price of food has the potential to improve health, reduce GHGEs, and raise revenue. The simple addition of a tax on SSBs can mitigate negative health consequences arising from sugar being low in GHGEs. Further conflicts remain, including increased consumption of unhealthy foods such as cakes and nutrients such as salt.

Classification of Parkinson’s Disease using MultiPass Lvq,Logistic Model Tree,K-Star for Audio Data set : Classification of Parkinson Disease using Audio Dataset

Parkinson's disease (PD) is a degenerative illness whose cardinal symptoms include rigidity, tremor, and slowness of movement. In addition to its widely recognized effects PD can have a profound effect on speech and voice.The speech symptoms most commonly demonstrated by patients with PD are reduced vocal loudness, monopitch, disruptions of voice quality, and abnormally fast rate of speech. This cluster of speech symptoms is often termed Hypokinetic Dysarthria.The disease can be difficult to diagnose accurately, especially in its early stages, due to this reason, automatic techniques based on Artificial Intelligence should increase the diagnosing accuracy and to help the doctors make better decisions. The aim of the thesis work is to predict the PD based on the audio files collected from various patients.Audio files are preprocessed in order to attain the features.The preprocessed data contains 23 attributes and 195 instances. On an average there are six voice recordings per person, By using data compression technique such as Discrete Cosine Transform (DCT) number of instances can be minimized, after data compression, attribute selection is done using several WEKA build in methods such as ChiSquared, GainRatio, Infogain after identifying the important attributes, we evaluate attributes one by one by using stepwise regression.Based on the selected attributes we process in WEKA by using cost sensitive classifier with various algorithms like MultiPass LVQ, Logistic Model Tree(LMT), K-Star.The classified results shows on an average 80%.By using this features 95% approximate classification of PD is acheived.This shows that using the audio dataset, PD could be predicted with a higher level of accuracy.