Estimation Of Model Coefficients Research Articles

Hemolytic adverse effects of intravenous immunoglobulin: modeling predicts risk reduction with anti-A/B immunoaffinity chromatography and to a lesser extent with anti-A donor screening.

The risk of hemolytic events (HEs) with intravenous immunoglobulin (IVIG) therapy appears to be linked to isoagglutinins (anti-A and anti-B) in the product. Patient risk factors include high IVIG dose, blood group, and underlying inflammatory state. Using published anti-A and anti-B titers for IVIG products and HE rates calculated from HEs spontaneously reported to EudraVigilance, regression models were developed to infer the relationship between HE risk and IVIG isoagglutinin levels for each blood group. Applying estimated model coefficients to isoagglutinin levels associated with an IVIG (Privigen; CSL Behring, King of Prussia, PA), manufactured with and without isoagglutinin reduction steps, predicted HE risk values were generated for each product: 1) without any isoagglutinin reduction, 2) anti-A donor screening, and 3) anti-A/anti-B specific immunoaffinity chromatography (IAC; Ig IsoLo). Predicted HE risk was highest for blood group AB, followed by A and B; it was low for O. Projected population shares of HEs by blood group were similar to reported real-world data. Compared with the original process (no isoagglutinin reduction), the model predicts lower hemolytic risk with anti-A donor screening and even lower hemolytic risk with IAC isoagglutinin reduction. IAC isoagglutinin reduction is predicted to reduce the HE risk with IVIG substantially. Physicians should be especially vigilant to HE risk in patients with blood group AB and, to a lesser extent, A when using IVIG products with high anti-A titers.

Method of indirect measurements on the basis of vector-observation

The method of indirect measurements is considered, which allows to increase the accuracy of estimation of the measured sought value by building a multiresponse model of measurements by means of vector-observation. The expressions for calculation of the model coefficient estimates and covariance matrices of measurement errors are given. It is shown, how by means of indirect measurements and the multiresponse model the refined estimation of the sought value and its confidence interval are calculated.

An Efficient Parametric Model-based Framework for Recursive Frequency/Spectrum Estimation of Nonstationary Signal

The manuscript intends to a design a general form of computationally efficient parametric mechanism based model to estimate the recursive frequency/spectrum and describe the nonlinear signals which consists of diverse degrees of nonlinearity and and indiscreet units. The time variant frequency estimation is defined as the as a time-varying model recognizable proof issue in which faulty/failure data are evaluated by model coefficients. In this, anestimation approach of QR-disintegration based recursive slightest M-gauge (QRRLM) is utilized for estimation of recursive time-vareint model coefficients in non-linear environment conditionby utilizing M-estimation. Here, a Veriable Forgetting Factor Control (VFFC) are designed to enhance the exection of QRRLM mechanism in nonlinear condition. In this, a hypothetical deduction and re-enactments approaches were used which helps to perform VFFC determination. The resultant VFFC-QRRLM estimation can confine and limit the faulty unitswhile dealing with different degrees of nonlinearvariations. Recreation comes about demonstrate that the proposed VFF-QRRLM calculation is more vigorous and exact than traditional recursive minimum squares-based techniques in evaluating both time-shifting narrowband recurrence segments and broadband otherworldly segments with incautious parts. Potential uses of the proposed technique can be found in quality force checking, online deficiency location, and discourse examination.

Using Gis-based Spatial Analysis To Determine Factors Influencing the Formation of Sinkholes in Greene County, Missouri

Abstract Sinkholes are inherent features of the karst terrain of Greene County, Missouri, that present hazards and engineering challenges to construction/infrastructure development. Analysis of relationships between the spatial distribution of sinkholes and possible influencing factors can help in understanding the controls involved in the formation of sinkholes. The spatial analysis outlined herein can aid in the assessment of potential sinkhole hazards. In this research, Geographic Information System–based ordinary least squares regression (OLS) and geographically weighted regression (GWR) methods were used to determine and evaluate principal factors appearing to influence the formation and distribution of karst sinkholes. From the OLS result, seven out of 12 possible influencing factors were found to exert significant control on sinkhole formation processes in the study area. These factors are overburden thickness, depth to groundwater, slope of the ground surface, distance to the nearest surface drainage line, distance to the nearest geological structure (such as faults or folds), distance to the nearest road, and distance to the nearest spring. These factors were then used as independent variables in the GWR model. The GWR model examined the spatial non-stationarity among the various factors and demonstrated better performance over OLS. GWR model coefficient estimates for each variable were mapped. These maps provide spatial insights into the influence of the variables on sinkhole densities throughout the study area. GWR spatial analysis appears to be an effective approach to understand sinkhole-influencing factors. The results could be useful to provide an objective means of parameter weighting in models of sinkhole susceptibility or hazard mapping.

A modified non-negative LMS algorithm for online system identification

Adaptive filters are useful solutions for system identification problem where an optimization problem is used to formulated the estimation of the unknown model coefficients. The nonnegativity constraint is one of the most frequently used constraint which can be imposed to avoid physically unreasonable solutions and to comply with physical characteristics. In this letter, we propose a new variant of non-negative least mean square (NNLMS) that has a less mean square error (MSE) value and faster convergence rate. We provide both mean weight behavior and transient excess mean-square error analysis for proposed algorithm. Simulation results validate the theoretical analysis and show the effectiveness of our proposed algorithm.

Orthogonal Scalar Feedback Digital Pre-Distortion Linearization

Digital pre-distortion (DPD) systems are employed to reduce intermodulation at the output of intrinsic non-linear RF power amplifiers. Typical DPD algorithms use time-domain linear regression to calculate the model coefficients used to compensate the amplifier’s distortion. This solution has high complexity and limited performance. This paper presents an innovative DPD system where the feedback information is a scalar measurement taken from the feedback signal. Different feedback information can be obtained from the distorted signal, e.g., the adjacent channel leakage rejection or the spectral mask margin. The DPD model coefficients estimation becomes a generic numerical optimization problem and is conducted iteratively in a coefficient-by-coefficient basis. A new coefficient orthogonalization algorithm removes the interaction between orders, resulting in a faster convergence time and lower output power variation across the iterations. The numerical optimization problem is simplified, since the nonlinear model becomes orthogonal in the intermodulation domain, i.e., the adjustment of given coefficient does not affect the others. Simulations and experimental results show that the proposed algorithm can effectively and efficiently compensate the nonlinear distortion and reduce the spectral regrowth.

The New York Sepsis Severity Score: Development of a Risk-Adjusted Severity Model for Sepsis.

In accordance with Rory's Regulations, hospitals across New York State developed and implemented protocols for sepsis recognition and treatment to reduce variations in evidence informed care and preventable mortality. The New York Department of Health sought to develop a risk assessment model for accurate and standardized hospital mortality comparisons of adult septic patients across institutions using case-mix adjustment. Retrospective evaluation of prospectively collected data. Data from 43,204 severe sepsis and septic shock patients from 179 hospitals across New York State were evaluated. Prospective data were submitted to a database from January 1, 2015, to December 31, 2015. None. Maximum likelihood logistic regression was used to estimate model coefficients used in the New York State risk model. The mortality probability was estimated using a logistic regression model. Variables to be included in the model were determined as part of the model-building process. Interactions between variables were included if they made clinical sense and if their p values were less than 0.05. Model development used a random sample of 90% of available patients and was validated using the remaining 10%. Hosmer-Lemeshow goodness of fit p values were considerably greater than 0.05, suggesting good calibration. Areas under the receiver operator curve in the developmental and validation subsets were 0.770 (95% CI, 0.765-0.775) and 0.773 (95% CI, 0.758-0.787), respectively, indicating good discrimination. Development and validation datasets had similar distributions of estimated mortality probabilities. Mortality increased with rising age, comorbidities, and lactate. The New York Sepsis Severity Score accurately estimated the probability of hospital mortality in severe sepsis and septic shock patients. It performed well with respect to calibration and discrimination. This sepsis-specific model provides an accurate, comprehensive method for standardized mortality comparison of adult patients with severe sepsis and septic shock.

Estimating Parameters in Autoregressive Models in Non-Normal Situations: Symmetric Innovations

The estimation of coefficients in a simple regression model with autocorrelated errors is considered. The underlying distribution is assumed to be symmetric, one of Student's t family for illustration. Closed form estimators are obtained and shown to be remarkably efficient and robust. Skew distributions will be considered in a future paper.

基于结构方程模型的大学生时间管理对学业影响的相关性分析

结构方程模型是一种分析建立、统计预测多变量因果关系的数学分析方法。本文从心理学的角度出发，将学业设为学习能力、创造性思维、语言表达能力、交流合作能力四个维度，将时间设为时间价值感、时间监控观和时间效能感三个维度。对长春市四所高校的学生发放700份调查问卷，使用SPSS统计软件进行问卷信度分析，得到问卷信度为0.893，表明信度良好。用AMOS软件建立模型系数估计统计量以及结构方程模型路径图，得到所有项路径系数的P值都在0.05的显著性水平以下，表明大学生时间管理和学业水平存在显著正相关，并据此建立了相关的结构方程。研究结果显示，好的时间管理不仅可以提高学生的学习绩效，还可以为促进人格的进一步社会化奠定良好的基础。 The structural equation model is a mathematical analysis method to establish and statistically predict the causality of multivariate variables. From the perspective of psychology, this article sets the academic as the four dimensions of learning ability, creative thinking, verbal expression ability and communication and cooperation ability, setting time as the three dimensions of time value, time monitoring and time efficacy. 700 questionnaires were distributed to students in four colleges and universities in Changchun, and the questionnaire reliability was analyzed by using SPSS statistical software. The reliability of the questionnaire was 0.893, indicating good reliability. Using the AMOS software to establish the model coefficient estimation statistic and the structural equation model path map, the P values of all the path coefficients are all below the significance level of 0.05, indicating that there is a significant positive correlation between the time management and the academic level of undergraduates and the related structural equa-tions are established. The results show that good time management can not only improve stu-dents' learning performance, but also lay a good foundation for further socialization of person-ality.

Turbulent transport and mixing in transitional Rayleigh-Taylor unstable flow: A priori assessment of gradient-diffusion and similarity modeling.

Data from a 1152×760×1280 direct numerical simulation [N. J. Mueschke and O. Schilling, Phys.Fluids 21, 014106 (2009)PHFLE61070-663110.1063/1.3064120] of a Rayleigh-Taylor mixing layer modeled after a small-Atwood-number water-channel experiment is used to investigate the validity of gradient diffusion and similarity closures a priori. The budgets of the mean flow, turbulent kinetic energy, turbulent kinetic energy dissipation rate, heavy-fluid mass fraction variance, and heavy-fluid mass fraction variance dissipation rate transport equations across the mixing layer were previously analyzed [O. Schilling and N. J. Mueschke, Phys. Fluids 22, 105102 (2010)PHFLE61070-663110.1063/1.3484247] at different evolution times to identify the most important transport and mixing mechanisms. Here a methodology is introduced to systematically estimate model coefficients as a function of time in the closures of the dynamically significant terms in the transport equations by minimizing the L_{2} norm of the difference between the model and correlations constructed using the simulation data. It is shown that gradient-diffusion and similarity closures used for the turbulent kinetic energy K, turbulent kinetic energy dissipation rate ε, heavy-fluid mass fraction variance S, and heavy-fluid mass fraction variance dissipation rate χ equations capture the shape of the exact, unclosed profiles well over the nonlinear and turbulent evolution regimes. Using order-of-magnitude estimates [O. Schilling and N. J. Mueschke, Phys.Fluids 22, 105102 (2010)PHFLE61070-663110.1063/1.3484247] for the terms in the exact transport equations and their closure models, it is shown that several of the standard closures for the turbulent production and dissipation (destruction) must be modified to include Reynolds-number scalings appropriate for Rayleigh-Taylor flow at small to intermediate Reynolds numbers. The late-time, large Reynolds number coefficients are determined to be different from those used in shear flow applications and largely adopted in two-equation Reynolds-averaged Navier-Stokes (RANS) models of Rayleigh-Taylor turbulent mixing. In addition, it is shown that the predictions of the Boussinesq model for the Reynolds stress agree better with the data when additional buoyancy-related terms are included. It is shown that an unsteady RANS paradigm is needed to predict the transitional flow dynamics from early evolution times, analogous to the small Reynolds number modifications in RANS models of wall-bounded flows in which the production-to-dissipation ratio is far from equilibrium. Although the present study is specific to one particular flow and one set of initial conditions, the methodology could be applied to calibrations of other Rayleigh-Taylor flows with different initial conditions (which may give different results during the early-time, transitional flow stages, and perhaps asymptotic stage). The implications of these findings for developing high-fidelity eddy viscosity-based turbulent transport and mixing models of Rayleigh-Taylor turbulence are discussed.

Shark length–length relationships: Studying morphology allows the detection of bias in routine fisheries sampling

Reliable information on length–length relationships is required for standardising and comparing size measures and for improving the use of available data and biological information collected as a function of different size measures. Here, we used biological information on sharks and rays collected by on-board observers in Western Australia since the 1990s to estimate the total length–fork length relationships for 31 species. For an additional five species, the estimated model coefficients were unreliable. Further analysis for four wobbegong species indicated that some records could have been misidentified. For sixgill sharks, the linear model fitted the data well but visual exploration showed that two observers systematically measured this species differently. Our study shows that length–length relationships can be used to detect biases in biological data sets.

Identification-based simplified model of large container ships using support vector machines and artificial bee colony algorithm

The 6 degrees of freedom (DOF) model with a high degree of complexity for capturing ship dynamics is generally able to track the nonlinear and coupling dynamics of ships. However, the 6 DOF model makes challenges in estimating model coefficients and designing the model-based control. Therefore, simplified ship dynamic models within allowed accuracy are essential. This paper simplified the 6 DOF nonlinear dynamic model of ships into two decoupled models including the speed model and the steering model through reasonable assumptions. Those models were tested through maneuvering simulations of a container ship with a 4 DOF dynamic model. Support vector machines (SVM) optimized by the artificial bee colony algorithm (ABC) was used to identify parameters of speed and steering models by analyzing the rudder angle, propeller shaft speed, surge and sway velocities, and yaw rate from simulated data extracted from a series of maneuvers made by the container ship. Comparisons with the first order linear and nonlinear Nomoto models show that the simplified nonlinear steering model can capture more complicated dynamics and performs better. Additionally, comparisons among three different parameter identification methods demonstrate similar identification results but the different performance involving the applicability and effectiveness. SVM optimized by ABC is relatively convenient and effective for parameter identification of ship simplified dynamic models.

Performance analysis of the preliminary test estimator with series of stochastic restrictions

ABSTRACTIn this paper, the problem of estimation of the regression coefficients in a multiple regression model is considered under the multicollinearity situation when there are series of stochastic linear restrictions available on the regression parameter vector. We have considered the preliminary test ridge regression estimators (PTRREs) based on the Wald, likelihood ratio, and lagrangian multiplier tests. Tables for the maximum and minimum guaranteed efficiency of the PTRREs are obtained, which allow us to determine the optimum choice of the level of significance corresponding to the optimum estimator. Some numerical results support the findings.

Enhancing Classification Performance of Functional Near-Infrared Spectroscopy- Brain-Computer Interface Using Adaptive Estimation of General Linear Model Coefficients.

In this paper, a novel methodology for enhanced classification of functional near-infrared spectroscopy (fNIRS) signals utilizable in a two-class [motor imagery (MI) and rest; mental rotation (MR) and rest] brain–computer interface (BCI) is presented. First, fNIRS signals corresponding to MI and MR are acquired from the motor and prefrontal cortex, respectively, afterward, filtered to remove physiological noises. Then, the signals are modeled using the general linear model, the coefficients of which are adaptively estimated using the least squares technique. Subsequently, multiple feature combinations of estimated coefficients were used for classification. The best classification accuracies achieved for five subjects, for MI versus rest are 79.5, 83.7, 82.6, 81.4, and 84.1% whereas those for MR versus rest are 85.5, 85.2, 87.8, 83.7, and 84.8%, respectively, using support vector machine. These results are compared with the best classification accuracies obtained using the conventional hemodynamic response. By means of the proposed methodology, the average classification accuracy obtained was significantly higher (p < 0.05). These results serve to demonstrate the feasibility of developing a high-classification-performance fNIRS-BCI.

Hybrid three-phase estimators for large-area forest inventory using ground plots, airborne lidar, and space lidar

Previous studies have utilized ground plots, airborne lidar scanning or profiling data, and space lidar profiling data to estimate biomass across large regions, but these studies have failed to take into account the variance components associated with multiple models because the proper variance equations were not available. Previous large-domain studies estimated the variances of their biomass density estimates as the sum of the GLAS sampling variability plus the model variability associated with the models that predict airborne lidar estimates of biomass density (Y) as a function of satellite lidar measurements (X). This approach ignores the additional variability associated with the predictive models used to estimate ground biomass density as a function of airborne lidar measurements. This paper addresses that shortcoming. Analytic variance expressions are provided that include sampling variability and model variability in situations where multiple models are employed to generate estimates of biomass. As an example, the forest biomass of the continental US is estimated, by forest stratum within state, using a space lidar system (ICESat/GLAS). An airborne laser system (ALS) is used as an intermediary to tie the GLAS measurements of forest height to a small subset of US Forest Service (USFS) ground plots by flying the ALS over the ground plots and, independently, over individual GLAS footprints. Two sets of models are employed to relate satellite measurements to the ground plots. The first set of equations relates USFS ground plot estimates of total aboveground dry biomass density (Y1) to spatially coincident ALS forest canopy measurements (X1). The second set of models predicts those ALS canopy height measurements (X1) used in the first set of models to GLAS waveform measurements (X2). The following important conclusions are noted. (1) The variability associated with estimation of the plot-ALS model coefficients is significant and should be included in the overall estimate of biomass density variance. In the continental US, the total variance of mean forest biomass density (98.06t/ha) increases by a factor of 3.6×, i.e., from 1.91 to 6.94t2/ha2, when plot-ALS model variance is included in the calculation of total variance. (2) State-level results are more variable, but on average, the percent model variance at the state level, i.e., (model variance/total variance)∗100, increases from 16% to 59% when plot-ALS model variance is included. (3) The overall model variance is driven in large part by the number of plots overflown by the ALS and the number of GLAS pulses overflown by the ALS. Given a choice of improving precision by either increasing the number of plot-ALS observations or increasing ALS-GLAS observations, there is no obvious benefit to selecting one over the other. However, typically the number of ground plots overflown is the limiting factor. (4) If heteroskedasticity is evident in either the ground-air or air-satellite models, it can modeled using weighted regression techniques and incorporated into these model variance formulas in straightforward fashion. The results are unambiguous; in a hybrid three-phase sampling framework, both the ground-air and air-satellite model variance components are significant and should be taken into account.

Generalized Additive Models for Gigadata: Modeling the U.K. Black Smoke Network Daily Data

We develop scalable methods for fitting penalized regression spline based generalized additive models with of the order of 104 coefficients to up to 108 data. Computational feasibility rests on: (i) a new iteration scheme for estimation of model coefficients and smoothing parameters, avoiding poorly scaling matrix operations; (ii) parallelization of the iteration’s pivoted block Cholesky and basic matrix operations; (iii) the marginal discretization of model covariates to reduce memory footprint, with efficient scalable methods for computing required crossproducts directly from the discrete representation. Marginal discretization enables much finer discretization than joint discretization would permit. We were motivated by the need to model four decades worth of daily particulate data from the U.K. Black Smoke and Sulphur Dioxide Monitoring Network. Although reduced in size recently, over 2000 stations have at some time been part of the network, resulting in some 10 million measurements. Modeling at a daily scale is desirable for accurate trend estimation and mapping, and to provide daily exposure estimates for epidemiological cohort studies. Because of the dataset size, previous work has focused on modeling time or space averaged pollution levels, but this is unsatisfactory from a health perspective, since it is often acute exposure locally and on the time scale of days that is of most importance in driving adverse health outcomes. If computed by conventional means our black smoke model would require a half terabyte of storage just for the model matrix, whereas we are able to compute with it on a desktop workstation. The best previously available reduced memory footprint method would have required three orders of magnitude more computing time than our new method. Supplementary materials for this article are available online.

Influence of inhomogeneous surface heat capacity on the estimation of radiative response coefficients in a two-zone energy balance model

Observationally constrained values of the global radiative response coefficient are pivotal to assess the reliability of modeled climate feedbacks. A widely used approach is to measure transient global radiative imbalance related to surface temperature changes. However, in this approach, a potential error in the estimate of radiative response coefficients may arise from surface inhomogeneity in the climate system. We examined this issue theoretically using a simple two-zone energy balance model. Here, we dealt with the potential error by subtracting the prescribed radiative response coefficient from those calculated within the two-zone framework. Each zone was characterized by the different magnitude of the radiative response coefficient and the surface heat capacity, and the dynamical heat transport in the atmosphere between the zones was parameterized as a linear function of the temperature difference between the zones. Then, the model system was forced by randomly generated monthly varying forcing mimicking time-varying forcing like an observation. The repeated simulations showed that inhomogeneous surface heat capacity causes considerable miscalculation (down to −1.4 W m−2 K−1 equivalent to 31.3% of the prescribed value) in the global radiative response coefficient. Also, the dynamical heat transport reduced this miscalculation driven by inhomogeneity of surface heat capacity. Therefore, the estimation of radiative response coefficients using the surface temperature-radiation relation is appropriate for homogeneous surface areas least affected by the exterior.

Development of hybrid artificial intelligent based handover decision algorithm

The possibility of seamless handover remains a mirage despite the plethora of existing handover algorithms. The underlying factor responsible for this has been traced to the Handover decision module in the Handover process. Hence, in this paper, the development of novel hybrid artificial intelligent handover decision algorithm has been developed. The developed model is made up of hybrid of Artificial Neural Network (ANN) based prediction model and Fuzzy Logic. On accessing the network, the Received Signal Strength (RSS) was acquired over a period of time to form a time series data. The data was then fed to the newly proposed k-step ahead ANN-based RSS prediction system for estimation of prediction model coefficients. The synaptic weights and adaptive coefficients of the trained ANN was then used to compute the k-step ahead ANN based RSS prediction model coefficients. The predicted RSS value was later codified as Fuzzy sets and in conjunction with other measured network parameters were fed into the Fuzzy logic controller in order to finalize handover decision process. The performance of the newly developed k-step ahead ANN based RSS prediction algorithm was evaluated using simulated and real data acquired from available mobile communication networks. Results obtained in both cases shows that the proposed algorithm is capable of predicting ahead the RSS value to about ±0.0002dB. Also, the cascaded effect of the complete handover decision module was also evaluated. Results obtained show that the newly proposed hybrid approach was able to reduce ping-pong effect associated with other handover techniques.

Abstract WP218: Brain Reserve: A Protective Mechanism for Stroke Outcome

Introduction: Stroke is a leading cause of disability worldwide. Mechanisms of post-stroke recovery are complex, and conventional outcome prediction models are limited. “Brain reserve” (BR) has been proposed as a construct to model the brain’s capacity to withstand insults. BR has been shown to co-vary with white matter hyperintensity volume (WMHv) using structural equation modeling (SEM), a technique to test models with latent variables. Hypothesis: We hypothesize that BR is a protective mechanism that improves functional outcome after acute ischemic stroke (AIS). Methods: We define an effective brain reserve (eBR), the remaining brain reserve after other influences have been accounted for. Using SEM, we characterize eBR through intra-cranial volume (ICV), age and systolic blood pressure (SBP). Our model incorporates known relationships between age, SBP, WMHv, acute infarct volume on diffusion-weighted imaging (DWIv) and 90-day functional post-stroke outcome (modified Rankin Scale; mRS), as shown in Figure 1. Path analysis was performed (R; package lavaan) to estimate the relations within the model in a dataset of 451 AIS patients. No priors were used for the path coefficients. Results: The estimated model coefficients (Figure 1) show that eBR is negatively associated with age and SBP, but positively with ICV, while association between age, SBP and WMHv are positive. Outcome is positively associated with WMHv and DWIv and negatively with eBR, suggesting that eBR acts as a protective mechanism. All path coefficients are statistically significant, except for WMHv and mRS. Conclusion: Our analysis shows that eBR is negatively associated with post-stroke outcome (the higher eBR, the lower mRS), suggesting that eBR acts as a protective mechanism. Additionally we reproduced known relationships between WMHv, SBP, age, DWIv and mRS.

Identification of parametric models with a priori knowledge of process properties

Abstract An approach to estimation of a parametric discrete-time model of a process in the case of some a priori knowledge of the investigated process properties is presented. The knowledge of plant properties is introduced in the form of linear bounds, which can be determined for the coefficient vector of the parametric model studied. The approach yields special biased estimation of model coefficients that preserves demanded properties. A formula for estimation of the model coefficients is derived and combined with a recursive scheme determined for minimization of the sum of absolute model errors. The estimation problem of a model with known static gains of inputs is discussed and proper formulas are derived. This approach can overcome the non-identifiability problem which has been observed during estimation based on measurements recorded in industrial closed-loop control systems. The application of the proposed approach to estimation of a model for an industrial plant (a water injector into the steam flow in a power plant) is presented and discussed.