A Comparison of Machine Learning Techniques for Modeling River Flow Time Series: The Case of Upper Cauvery River Basin

Models of river flow time series are essential in efficient management of a river basin. It helps policy makers in developing efficient water utilization strategies to maximize the utility of scarce water resource. Time series analysis has been used extensively for modeling river flow data. The use of machine learning techniques such as support-vector regression and neural network models is gaining increasing popularity. In this paper we compare the performance of these techniques by applying it to a long-term time-series data of the inflows into the Krishnaraja Sagar reservoir (KRS) from three tributaries of the river Cauvery. In this study flow data over a period of 30 years from three different observation points established in upper Cauvery river sub-basin is analyzed to estimate their contribution to KRS. Specifically, ANN model uses a multi-layer feed forward network trained with a back-propagation algorithm and support vector regression with epsilon intensive-loss function is used. Auto-regressive moving average models are also applied to the same data. The performance of different techniques is compared using performance metrics such as root mean squared error (RMSE), correlation, normalized root mean squared error (NRMSE) and Nash-Sutcliffe Efficiency (NSE).

A Hierarchical Model for Ordinal Matrix Factorization

This paper proposes a hierarchical probabilistic model for ordinal matrix factorization. Unlike previous approaches, we model the ordinal nature of the data and take a principled approach to incorporating priors for the hidden variables. Two algorithms are presented for inference, one based on Gibbs sampling and one based on variational Bayes. Importantly, these algorithms may be implemented in the factorization of very large matrices with missing entries. The model is evaluated on a collaborative filtering task, where users have rated a collection of movies and the system is asked to predict their ratings for other movies. The Netflix data set is used for evaluation, which consists of around 100 million ratings. Using root mean-squared error (RMSE) as an evaluation metric, results show that the suggested model outperforms alternative factorization techniques. Results also show how Gibbs sampling outperforms variational Bayes on this task, despite the large number of ratings and model parameters. Matlab implementations of the proposed algorithms are available from cogsys.imm.dtu.dk/ordinalmatrixfactorization.

Forecasting overseas visitors into the United Kingdom using continuous time and autoregressive fractional integrated moving average models with discrete data

This paper applies Gaussian estimation methods to continuous time models for modelling overseas visitors into the UK. The use of continuous time modelling is widely used in economics and finance but not in tourism forecasting. Using monthly data for 1986–2010, various continuous time models are estimated and compared to autoregressive integrated moving average (ARIMA) and autoregressive fractionally integrated moving average (ARFIMA) models. Dynamic forecasts are obtained over different periods. The empirical results show that the ARIMA model performs very well, but that the constant elasticity of variance (CEV) continuous time model has the lowest root mean squared error (RMSE) over a short period.

Evaluation of the realism of climate reconstruction using the Coexistence Approach with modern pollen samples from the Tibetan Plateau

We apply the Coexistence Approach (CoA) to reconstruct mean annual precipitation (MAP), mean annual temperature (MAT), mean temperature of thewarmestmonth (MTWA) and mean temperature of the coldest month (MTCO) at 44 pollen sites on the Qinghai–Tibetan Plateau. The modern climate ranges of the taxa are obtained (1) from county-level presence/absence data and (2) from data on the optimum and range of each taxon from Lu et al. (2011). The CoA based on the optimumand range data yields better predictions of observed climate parameters at the pollen sites than that based on the county-level data. The presence of arboreal pollen, most of which is derived fromoutside the region, distorts the reconstructions. More reliable reconstructions are obtained using only the non-arboreal component of the pollen assemblages. The root mean-squared error (RMSE) of the MAP reconstructions are smaller than the RMSE of MAT, MTWA and MTCO, suggesting that precipitation gradients are the most important control of vegetation distribution on the Qinghai–Tibetan Plateau. Our results show that CoA could be used to reconstruct past climates in this region, although in areas characterized by open vegetation the most reliable estimates will be obtained by excluding possible arboreal contaminants.

Convex combinations of long memory estimates from different sampling rates

Convex combinations of long memory estimates using the same data observed at different sampling rates can decrease the standard deviation of the estimates, at the cost of inducing a slight bias. The convex combination of such estimates requires a preliminary correction for the bias observed at lower sampling rates, reported by Souza and Smith (2002). Through Monte Carlo simulations, we investigate the bias and the standard deviation of the combined estimates, as well as the root mean squared error (RMSE), which takes both into account. While comparing the results of standard methods and their combined versions, the latter achieve lower RMSE, for the two semi-parametric estimators under study (by about 30% on average for ARFIMA(0,d,0) series).

Towards the construction of a validated numerical system to study the mesoscale dynamics of the North East Atlantic (2003-2006)

Máster en Oceanografía

Sistematización de suelos : eficiencia en la ejecución del proyecto en suelos complejos

El manejo sustentable de los recursos naturales relacionados con proyectos de utilización de los recursos hídricos (entre otros), requiere en muchos casos de la modificación del relieve existente. Esto conlleva la necesidad de adecuación de la capa homogénea superior del suelo, operación que suele denominarse "sistematización", la cual facilita una distribución más uniforme de las lluvias y del agua de riego. Esta modificación de la capa superior del suelo es realizada en base a un proyecto, cuya inclinación responda a las pendientes naturales o a las establecidas por el diseñador. En la ejecución del diseño proyectado, en superficies superiores a una hectárea, el movimiento de tierra se realiza con equipos pesados, que no aseguran un alto porcentaje de eficiencia en lo que al movimiento de tierra se refiere, ya que parte del material se pierde en el acarreo, pero muy especialmente, por la compactación desuniforme del mismo, asociada con las texturas complejas del suelo a trabajar. El presente trabajo determinó el índice de precisión en la ejecución del proyecto de sistematización a partir de un índice estadístico internacionalmente aceptado, el "Root Mean Squared Error (RMSE)", comparando los valores altimétricos proyectados y los realmente obtenidos luego de la ejecución del proyecto, en tres parcelas con distinta secuencia de labores y maquinaria utilizadas, pero con el mismo tipo de suelo en el área del eje Pilar - La Plata (Argentina). Los resultados obtenidos, que varían de un RMSE de 4 a 6 cm, permiten concluir, para los sitios y las condiciones estudiadas, que no pueden asegurarse en la sistematización índices de precisión en la ejecución de la obra, inferiores a los 4 cm.

Inference of binding affinity from neuronal receptors in humans

Tese de mestrado, Bioinformática e Biologia Computacional (Bioinformática), Universidade de Lisboa, Faculdade de Ciências, 2016

Disentangling the impact of seroconversion age and set-point viral load on ART-free HIV survival

Thesis (Master's)--University of Washington, 2016-06

Mitigating congestion by integrating time forecasting and realtime information aggregation in cellular networks

An iterative travel time forecasting scheme, named the Advanced Multilane Prediction based Real-time Fastest Path (AMPRFP) algorithm, is presented in this dissertation. This scheme is derived from the conventional kernel estimator based prediction model by the association of real-time nonlinear impacts that caused by neighboring arcs’ traffic patterns with the historical traffic behaviors. The AMPRFP algorithm is evaluated by prediction of the travel time of congested arcs in the urban area of Jacksonville City. Experiment results illustrate that the proposed scheme is able to significantly reduce both the relative mean error (RME) and the root-mean-squared error (RMSE) of the predicted travel time. To obtain high quality real-time traffic information, which is essential to the performance of the AMPRFP algorithm, a data clean scheme enhanced empirical learning (DCSEEL) algorithm is also introduced. This novel method investigates the correlation between distance and direction in the geometrical map, which is not considered in existing fingerprint localization methods. Specifically, empirical learning methods are applied to minimize the error that exists in the estimated distance. A direction filter is developed to clean joints that have negative influence to the localization accuracy. Synthetic experiments in urban, suburban and rural environments are designed to evaluate the performance of DCSEEL algorithm in determining the cellular probe’s position. The results show that the cellular probe’s localization accuracy can be notably improved by the DCSEEL algorithm. Additionally, a new fast correlation technique for overcoming the time efficiency problem of the existing correlation algorithm based floating car data (FCD) technique is developed. The matching process is transformed into a 1-dimensional (1-D) curve matching problem and the Fast Normalized Cross-Correlation (FNCC) algorithm is introduced to supersede the Pearson product Moment Correlation Co-efficient (PMCC) algorithm in order to achieve the real-time requirement of the FCD method. The fast correlation technique shows a significant improvement in reducing the computational cost without affecting the accuracy of the matching process.

SVR-GARCH com misturas de kernels gaussianos

Dissertação (mestrado)—Universidade de Brasília, Departamento de Administração, Programa de Pós-graduação em Administração, 2016.

Assessment of the spatial variability in Tall wheatgrass forage using LANDSAT 8 satellite imagery to delineate potential management zones

Little information is available on the degree of within-field variability of potential production of Tall wheatgrass (Thinopyrum ponticum) forage under unirrigated conditions. The aim of this study was to characterize the spatial variability of the accumulated biomass (AB) without nutritional limitations through vegetation indexes, and then use this information to determine potential management zones. A 27-×-27-m grid cell size was chosen and 84 biomass sampling areas (BSA), each 2 m(2) in size, were georeferenced. Nitrogen and phosphorus fertilizers were applied after an initial cut at 3 cm height. At 500 °C day, the AB from each sampling area, was collected and evaluated. The spatial variability of AB was estimated more accurately using the Normalized Difference Vegetation Index (NDVI), calculated from LANDSAT 8 images obtained on 24 November 2014 (NDVInov) and 10 December 2014 (NDVIdec) because the potential AB was highly associated with NDVInov and NDVIdec (r (2) = 0.85 and 0.83, respectively). These models between the potential AB data and NDVI were evaluated by root mean squared error (RMSE) and relative root mean squared error (RRMSE). This last coefficient was 12 and 15 % for NDVInov and NDVIdec, respectively. Potential AB and NDVI spatial correlation were quantified with semivariograms. The spatial dependence of AB was low. Six classes of NDVI were analyzed for comparison, and two management zones (MZ) were established with them. In order to evaluate if the NDVI method allows us to delimit MZ with different attainable yields, the AB estimated for these MZ were compared through an ANOVA test. The potential AB had significant differences among MZ. Based on these findings, it can be concluded that NDVI obtained from LANDSAT 8 images can be reliably used for creating MZ in soils under permanent pastures dominated by Tall wheatgrass.

Assessment of body composition in Sri Lankan children: Validation of a bioelectrical impedance prediction equation

Objective: To develop bioelectrical impedance analysis (BIA) equations to predict total body water (TBW) and fat-free mass (FFM) of Sri Lankan children. Subjects/Methods: Data were collected from 5- to 15-year-old healthy children. They were randomly assigned to validation (M/F: 105/83) and cross-validation (M/F: 53/41) groups. Height, weight and BIA were measured. TBW was assessed using isotope dilution method (D2 O). Multiple regression analysis was used to develop preliminary equations and cross-validated on an independent group. Final prediction equation was constructed combining the two groups and validated by PRESS (prediction of sum of squares) statistics. Impedance index (height2/impedance; cm2/Ω), weight and sex code (male = 1; female = 0) were used as variables. Results: Independent variables of the final prediction equation for TBW were able to predict 86.3% of variance with root means-squared error (RMSE) of 2.1l. PRESS statistics was 2.1l with press residuals of 1.2l. Independent variables were able to predict 86.9% of variance of FFM with RMSE of 2.7 kg. PRESS statistics was 2.8 kg with press residuals of 1.4 kg. Bland Altman technique showed that the majority of the residuals were within mean bias±1.96 s.d. Conclusions: Results of this study provide BIA equation for the prediction of TBW and FFM in Sri Lankan children. To the best of our knowledge there are no published BIA prediction equations validated on South Asian populations. Results of this study need to be affirmed by more studies on other closely related populations by using multi-component body composition assessment.

DISTRIBUTION FUNCTIONS, MEAN SQUARED ERRORS, AND CONFIDENCE LIMITS FOR RIDGE REGRESSION ESTIMATORS

A large number of ridge regression estimators have been proposed and used with little knowledge of their true distributions. Because of this lack of knowledge, these estimators cannot be used to test hypotheses or to form confidence intervals.^ This paper presents a basic technique for deriving the exact distribution functions for a class of generalized ridge estimators. The technique is applied to five prominent generalized ridge estimators. Graphs of the resulting distribution functions are presented. The actual behavior of these estimators is found to be considerably different than the behavior which is generally assumed for ridge estimators.^ This paper also uses the derived distributions to examine the mean squared error properties of the estimators. A technique for developing confidence intervals based on the generalized ridge estimators is also presented. ^