Statistical Parametric Model Research Articles

A statistical–parametric model of tropical cyclones for hazard assessment

Abstract. We present the formulation of an open-source, statistical–parametric model of tropical cyclones (TCs) for use in hazard and risk assessment applications. The model derives statistical relations for TC behaviour (genesis rate and location, intensity, speed and direction of translation) from best-track datasets, then uses these relations to create a synthetic catalogue based on stochastic sampling, representing many thousands of years of activity. A parametric wind field, based on radial profiles and boundary layer models, is applied to each event in the catalogue that is then used to fit extreme-value distributions for evaluation of return period wind speeds. We demonstrate the capability of the model to replicate observed behaviour of TCs, including coastal landfall rates which are of significant importance for risk assessments.

Model exploration using conditional visualization

AbstractIdeally, statistical parametric model fitting is followed by various summary tables which show predictor contributions, visualizations which assess model assumptions and goodness of fit, and test statistics which compare models. In contrast, modern machine‐learning fits are usually black box in nature, offer high‐performing predictions but suffer from an interpretability deficit. We examine how the paradigm of conditional visualization can be used to address this, specifically to explain predictor contributions, assess goodness of fit, and compare multiple, competing fits. We compare visualizations from techniques including trellis, condvis, visreg, lime, partial dependence, and ice plots. Our examples use random forest fits, but all techniques presented are model agnostic.This article is categorized under: Statistical and Graphical Methods of Data Analysis > Statistical Graphics and Visualization Statistical Learning and Exploratory Methods of the Data Sciences > Exploratory Data Analysis Statistical Learning and Exploratory Methods of the Data Sciences > Modeling Methods

A New Power Topp–Leone Generated Family of Distributions with Applications

In this paper, we introduce a new general family of distributions obtained by a subtle combination of two well-established families of distributions: the so-called power Topp–Leone-G and inverse exponential-G families. Its definition is centered around an original cumulative distribution function involving exponential and polynomial functions. Some desirable theoretical properties of the new family are discussed in full generality, with comprehensive results on stochastic ordering, quantile function and related measures, general moments and related measures, and the Shannon entropy. Then, a statistical parametric model is constructed from a special member of the family, defined with the use of the inverse Lomax distribution as the baseline distribution. The maximum likelihood method was applied to estimate the unknown model parameters. From the general theory of this method, the asymptotic confidence intervals of these parameters were deduced. A simulation study was conducted to evaluate the numerical behavior of the estimates we obtained. Finally, in order to highlight the practical perspectives of the new family, two real-life data sets were analyzed. All the measures considered are favorable to the new model in comparison to four serious competitors.

Statistical feature bag based background subtraction for local change detection

This article proposes a novel background subtraction (BGS) technique to detect local changes corresponding to the movement of the objects in video scenes. Here we propose an efficient combination of six local features; three existing and three newly proposed. For background modeling and subtraction here a statistical parametric biunique model is proposed. In the proposed BGS scheme, during the background training phase, the multi-valued features corresponding to background pixels are collected. A few simple statistical parameters are used to characterize each feature. For background subtraction, the multi-valued features computed at each pixel location are compared with those of the computed parameters corresponding to that feature. For each pixel location, different labels (either object or background) are obtained due to different features. For assigning a final label to the pixel in the target frame a majority voting based label fusion technique is used. The proposed technique is successfully tested over several video sequences and found to be providing better results compared to various existing state-of-the-art techniques with three performance evaluation measures.

HMM-Based Vietnamese Speech Synthesis

In this paper, improving naturalness HMM-based speech synthesis for Vietnamese language is described. By this synthesis method, trajectories of speech parameters are generated from the trained Hidden Markov models. A final speech waveform is synthesized from those speech parameters. The main objective for the development is to achieve maximum naturalness in output speech through key points. Firstly, system uses a high quality recorded Vietnamese speech database appropriate for training, especially in statistical parametric model approach. Secondly, prosodic informations such as tone, POS (part of speech) and features based on characteristics of Vietnamese language are added to ensure the quality of synthetic speech. Third, system uses STRAIGHT which showed its ability to produce high-quality voice manipulation and was successfully incorporated into HMM-based speech synthesis. The results collected show that the speech produced by our system has the best result when being compared with the other Vietnamese TTS systems trained from the same speech data.

Bayesian Networks for the Age Classification of Living Individuals: A Study on Transition Analysis

Over the past few decades, age estimation of living persons has represented a challenging task for many forensic services worldwide. In general, the process for age estimation includes the observation of the degree of maturity reached by some physical attributes, such as dentition or several ossification centers. The estimated chronological age or the probability that an individual belongs to a meaningful class of ages is then obtained from the observed degree of maturity by means of various statistical methods. Among these methods, those developed in a Bayesian framework offer to users the possibility of coherently dealing with the uncertainty associated with age estimation and of assessing in a transparent and logical way the probability that an examined individual is younger or older than a given age threshold. Recently, a Bayesian network for age estimation has been presented in scientific literature; this kind of probabilistic graphical tool may facilitate the use of the probabilistic approach. Probabilities of interest in the network are assigned by means of transition analysis, a statistical parametric model, which links the chronological age and the degree of maturity by means of specific regression models, such as logit or probit models. Since different regression models can be employed in transition analysis, the aim of this paper is to study the influence of the model in the classification of individuals. The analysis was performed using a dataset related to the ossifications status of the medial clavicular epiphysis and results support that the classification of individuals is not dependent on the choice of the regression model.

Theoretical and Computational Research of Heat Radiant Transfer in Cylinder

In order to obtain the radiation heat transfer theory and calculation methods, the movement of the gas particles, location, intensity, temperature, are researched in cylinder under different conditions with combustion system and the mode of heat transfer. Under high temperature conditions in the cylinder, the gas radiation heat transfer is researched in the complex heat transfer theory. A statistical correlation K narrow band model, a mean absorption coefficient, a gas line databases, by-line calculation method are found, through research and analysis emissivity, transmittance, absorption coefficient, typical models, mathematical equations, database, calculation methods. Examine the distribution performance of each database for different media concentration and temperature, a statistical narrow-band band parametric model accuracy is tested, using statistical narrow band model, the results of the use of by-line method. Research shows: selected spectral database, calculation method has a greater impact on the results. The research also shows the result coincides calculations based by-line HITEWP2010 database method, whether radiant heat or wall flux, statistical narrow band model. These are supplied to the internal combustion engine cylinder design.

A New Method of Face Modeling with Construction of Local Texture Model

The active shape model (ASM) is a statistical parametric model, which is mainly used in image feature extraction. On the basis of the analysis of the original ASM texture model, a new texture modeling method was proposed in this paper. The improved method fully utilized the gray level information of adjacent points in the neighborhood of sampling points, improved the original face modeling method which only used the one-dimensional gray information with model matching accuracy problems. The experiments on Weizmann face database Indicate that, the improved method can obviously improve the feature point positioning accuracy, and accelerate the speed of face model fitting.

3D reconstruction of rib cage geometry from biplanar radiographs using a statistical parametric model approach

Rib cage 3D reconstruction is an important prerequisite for thoracic spine modelling, particularly for studies of the deformed thorax in adolescent idiopathic scoliosis. This study proposes a new method for rib cage 3D reconstruction from biplanar radiographs, using a statistical parametric model approach. Simplified parametric models were defined at the hierarchical levels of rib cage surface, rib midline and rib surface, and applied on a database of 86 trunks. The resulting parameter database served to train statistical models which were used to quickly provide a first estimate of the reconstruction from identifications on both radiographs. This solution was then refined by manual adjustments in order to improve the matching between model and image. Accuracy was assessed by comparison with 29 rib cages from CT scans in terms of geometrical parameter differences and in terms of line-to-line error distance between the rib midlines. Intra and inter-observer reproducibility was determined for 20 scoliotic patients. The first estimate (mean reconstruction time of 2 min 30 s) was sufficient to extract the main rib cage global parameters with a 95% confidence interval lower than 7%, 8%, 2% and 4° for rib cage volume, antero-posterior and lateral maximal diameters and maximal rib hump, respectively. The mean error distance was 5.4 mm (max 35 mm) down to 3.6 mm (max 24 mm) after the manual adjustment step (3 min 30 s). The proposed method will improve developments of rib cage finite element modelling and evaluation of clinical outcomes.

Combination of supervised and unsupervised learning for training the activation functions of neural networks

Standard feedforward neural networks benefit from the nice theoretical properties of mixtures of sigmoid activation functions, but they may fail in several practical learning tasks. These tasks would be better faced by relying on a more appropriate, problem-specific basis of activation functions. The paper presents a connectionist model which exploits adaptive activation functions. Each hidden unit in the network is associated with a specific pair (f(·),p(·)), where f(·) is the activation function and p(·) is the likelihood of the unit being relevant to the computation of the network output over the current input. The function f(·) is optimized in a supervised manner, while p(·) is realized via a statistical parametric model learned through unsupervised (or, partially supervised) estimation. Since f(·) and p(·) influence each other’s learning process, the overall machine is implicitly a co-trained coupled model and, in turn, a flexible, non-standard neural architecture. Feasibility of the approach is corroborated by empirical evidence yielded by computer simulations involving regression and classification tasks.

Rotation invariant texture characterization using a curvelet based descriptor

This paper introduces a highly discriminative, precise and simple descriptor of natural textures, based on the curvelet transform. The proposed descriptor is calculated from the statistical pattern of the curvelet coefficients. The image is mapped to the curvelet space, where a statistical parametric model approaches the data distribution for each of the sub-bands. Once these parameters are estimated, they are subband-energy sorted out, achieving thereby the invariance to planar rotations. Finally, the Kullback–Leibler divergence between the statistical parameters is used to estimate a distance between images. We demonstrated the effectiveness of the proposed descriptor for classification and retrieval tasks.

Statistical M-Estimation and Consistency in Large Deformable Models for Image Warping

The problem of defining appropriate distances between shapes or images and modeling the variability of natural images by group transformations is at the heart of modern image analysis. A current trend is the study of probabilistic and statistical aspects of deformation models, and the development of consistent statistical procedure for the estimation of template images. In this paper, we consider a set of images randomly warped from a mean template which has to be recovered. For this, we define an appropriate statistical parametric model to generate random diffeomorphic deformations in two-dimensions. Then, we focus on the problem of estimating the mean pattern when the images are observed with noise. This problem is challenging both from a theoretical and a practical point of view. M-estimation theory enables us to build an estimator defined as a minimizer of a well-tailored empirical criterion. We prove the convergence of this estimator and propose a gradient descent algorithm to compute this M-estimator in practice. Simulations of template extraction and an application to image clustering and classification are also provided.

Statistical Tropical Cyclone Wind Radii Prediction Using Climatology and Persistence

Abstract An operational model used to predict tropical cyclone wind structure in terms of significant wind radii (i.e., 34-, 50-, and 64-kt wind radii, where 1 kt = 0.52 m s−1) at the National Oceanic and Atmospheric Administration/National Hurricane Center (NHC) and the Department of Defense/Joint Typhoon Warning Center (JTWC) is described. The statistical-parametric model employs aspects of climatology and persistence to forecast tropical cyclone wind radii through 5 days. Separate versions of the model are created for the Atlantic, east Pacific, and western North Pacific by statistically fitting a modified Rankine vortex, which is generalized to allow wavenumber-1 asymmetries, to observed values of tropical cyclone wind radii as reported by NHC and JTWC. Descriptions of the developmental data and methods used to formulate the model are given. A 2-yr verification and comparison with operational forecasts and an independently developed wind radii forecast method that also employs climatology and persistence suggests that the statistical-parametric model does a good job of forecasting wind radii. The statistical-parametric model also provides reliable operational forecasts that serve as a baseline for evaluating the skill of operational forecasts and other wind radii forecast methods in these tropical cyclone basins.

Non parametric space-time modeling of SO2 in presence of many missing data

Given pollution measurement from a network of monitoring sites in the area of a city and over an extended period of time, an important problem is to identify the spatial and temporal structure of the data. In this paper we focus on the identification and estimate of a statistical non parametric model to analyse the SO2 in the city of Padua, where data are collected by some fixed stations and some mobile stations moving without any specific rule in different new locations. The impact of the use of mobile stations is that for each location there are times when data was not collected. Assuming temporal stationarity and spatial isotropy for the residuals of an additive model for the logarithm of SO2 concentration, we estimate the semivariogram using a kernel-type estimator. Attempts are made to avoid the assumption of spatial isotropy. Bootstrap confidence bands are obtained for the spatial component of the additive model that is a deterministic function which defines the spatial structure. Finally, an example is proposed to design an optimal network for the mobiles monitoring stations in a fixed future time, given all the information available.

A new statistical model of tumor latency time

A new statistical parametric model of carcinogenesis is proposed. The focus is on tumor latency times. The model makes use of extreme value theory following the line initially explored by Pike [1] and further developed by Klebanov et al. [2]. In this paper, we make the first attempt to take account of the interdependence between the cells in an initiated tissue developing into a malignant tumor.

Prediction of cellular characteristics for various urban environments

The propagation channel for UHF/X-band waves in the city is modeled for two typical cases in the urban scene: (a) by regularly distributed rows of buildings placed on a flat terrain; and (b) by an array of randomly distributed buildings placed on rough terrain. The law of distribution of buildings in both cases is assumed to be Poisson. The loss characteristics in such urban propagation channels, as well as the co-channel interference parameter, the carrier-to-interference ratio (C/I), are investigated. In case (a), the three-dimensional multi-slit waveguide model is used for LOS conditions, and the two-dimensional multi-diffraction model is used for NLOS conditions. In case (b), the statistical parametric model of wave propagation is used, including single and multiple scattering effects, diffraction from buildings' roofs, the actual built-up relief, and various positions of receiver and transmitter antennas on rough terrain. The full algorithm for predicting propagation and cellular characteristics to increase the accuracy of radio and cellular maps is presented.

A model of ultrasound backscatter for the assessment of myocardial tissue structure and architecture.

A statistical parametric model of returning echoes from myocardium is theorized in order to investigate the relationship between normal myocardium structure and spectral signatures with the use of ultrasonic tissue characterization. It is hypothesized, that in a clinical setting the normal myofiber architecture in the left ventricular wall is structured as a matrix of cylinder scatterers whose orientation and spatial distribution vary according to two different statistical distribution laws: 1) a Gaussian law to approximate parametric angular myofiber variability at each site within the myocardial wall; 2) a gamma distribution law to describe parametric regularity in scatterer interdistance. In the model, the effect of the angle of insonification with respect to the alignment of myofibers on ultrasound backscatter was considered. The slope of the power spectral density (PSD) evaluated within the echocardiographic transducer bandwidth has been used as a ultrasonic tissue characterization parameter. The model has been tested by computer simulation and in vitro measurements on myocardial pig tissue specimens. The concordance between experimental and simulated results confirms that the model accounts for the process underlying the echo formation from normal myocardium. Moreover, it provides a simple method of simulation which can be easily implemented and used for the assessment of pathologic alterations.

Inversion for heat anomaly transport from sea surface temperature time series in the northwest Pacific

We describe a heat anomaly transport in the upper ocean mixed layer in the Kuroshio extension region and the subtropical gyre of the northwest Pacific. Emphasis is on behavior in the cool season (December‐March) during the Asian Winter Monsoon. The heat anomaly transport is estimated by applying an inversion technique to the stochastic partial differential equation for the heat anomaly balance of advection, diffusion, stabilizing feedback, and atmospheric forcing. The inversion consists of (1) derivation of statistical parametric model from the heat anomaly balance equation; (2) fitting the derived statistical model to the sea surface temperature (SST) anomaly covariances; and (3) calculation of the heat anomaly net advection velocity, horizontal diffusion coefficient, feedback factor and atmospheric forcing correlation from the parameters of the evaluated statistical model. The inversion was applied to the Comprehensive Ocean‐Atmosphere Data Set Compressed Marine Reports SST data averaged at 1° latitude × 2° longitude boxes on a 10‐day mean basis from 1965 to 1990. The estimates of the net advection velocity are consistent in magnitude and direction with the general circulation in the surface layer of the Northwest Pacific in winter. SST anomalies are transported to the west at ∼0.15 m s−1 in the northern part of the North Equatorial Current. Between 21° and 29°N in the recirculating region, SST anomalies propagate westward with the mean velocity less than 0.1 m s−1. South and east of Honshu the observed pattern of the SST anomaly transport agrees broadly with the circulations of the Kuroshio current and its extension and the Oyashio current. South of Honshu, the eastward transport is about 200–300 km wide; its absolute velocity is up to 0.2 m s−1. One branch of the transport separates from the coast near the large meander path of the Kuroshio current and follows the east‐southeast direction. The second separation from the coast occurs south of Hokkaido. Over the analysis domain the estimates of the diffusion coefficient are in the range of 3×103 to 6×103 m2 s−1. The higher values of the diffusion coefficient confirm the enhancement of the mesoscale eddy processes near the subtropical convergence zone. The analysis supports Hasselmann's (1976) theory in which generation of midlatitude SST anomalies lasting the dominant timescale of atmospheric processes is primarily attributed to the short period stochastic weather forcing. However, the analysis indicates that the inertia of SST anomalies to their “memory” of earlier winds can not be neglected in the vicinity of the western boundary and in the tropics.

Waiting time to pregnancy and pregnancy outcome among Danish workers in the textile, clothing, and footwear industries.

The relationship between time from planned to achieved pregnancy and pregnancy outcome has been studied in a group of 18,658 workers in the textile, clothing and footwear industries. Information on pregnancy outcome and delay in conception in the period 1979-84 was collected by self administered questionnaires in 1985. The response rate was 70.3%. During the study period there had been 5,171 live births and 708 spontaneous abortions. Information on delay in conception was collected in broad categories. The data were analysed by means of a newly developed statistical parametric model in order to collect all possible information from the highly grouped data. Median waiting time before a pregnancy which ended in spontaneous abortion was 1.68 times longer than median waiting time before a pregnancy leading to a live birth. There seems to be a correlation between the length of the waiting time and abortion.

Graphical tools for life time data analysis

It is purpose of this paper to introduce and describe graphical tools and techniques for (technically oriented) life time data analysis. The aim of this analysis is the determination of a representative statistical parametric or non-parametric model reflecting ageing behaviour over time. The paper intends to demonstrate that a subjective data analysis by graphs can be more sensitive and informative than pure parameter estimation in a predefined model or application of test statistics. In this sense some of those graphical tools are discussed which provide a less formal goodness-of-fit approach analysing the adherence to or deviation from a hypothesized distribution shape in a graph. Various graphical tools will be described and compared concerning their specific advantages and restrictions for life time data analysis. Finally some examples will demonstrate a “combined graphical analysis approach”.