• Journal of the Korean Statistical Society
• /
• 제29권2호
• /
• pp.187-199
• /
• 2000

In this paper, we deal with the asymptotic properties of the least absolute deviation estimators in the nonlinear time series regression model. For the sinusodial model which frequently appears in a time series analysis, we study the strong consistency and asymptotic normality of least absolute deviation estimators. And using the derived limiting distributions we show that the least absolute deviation estimators is more efficient than the least squared estimators when the error distribution of the model has heavy tails.

• Journal of the Korean Statistical Society
• /
• 제24권2호
• /
• pp.303-312
• /
• 1995

We consider the following type of general semi-parametric non-linear regression model : $y_i = f_i(\theta) + \epsilon_i, i=1, \cdots, n$ where ${f_i(\cdot)}$ represents the set of non-linear functions of the unknown parameter vector $\theta' = (\theta_1, \cdots, \theta_p)$ and ${\epsilon_i}$ represents the set of measurement errors with unknown distribution. Under suitable finite-sample, small-dispersion asymptotic framework, we derive a general lower bound for the asymptotic mean squared error (AMSE) matrix of the Gauss-consistent estimator of $\theta$. We then prove the fundamental result that the general non-linear least squares estimator (NLSE) is an optimal estimator within the class of all regular Gauss-consistent estimators irrespective of the type of the distribution of the measurement errors.

• The Journal of Asian Finance, Economics and Business
• /
• 제6권4호
• /
• pp.53-63
• /
• 2019

The paper examines the influence of profitability on distance to default (DD) in Vietnam securities market. The investigated sample consists of 211 companies listed on HOSE during 18 years from 2010 to 2017. We apply KMV model to calculate distance to default and use both macroeconomics factors and firm specific factors as independent variables. Using General Least Squared (GLS) method, we find evidence to confirm the positive relationship between profitability and distance to default. This result showed that, although profitability did not directly reflect the cash flow generated, a good profitable enterprise would be an important factor to help facilitate and generate cash flow and at the same time debt was guaranteed when it was due. Besides, the test results revealed that the financial structure and sales on assets have the inverse effect on the distance to default at the significance level of 5%. The results also revealed that a group of macro factors had an influence on the distance to default of businesses, including spread, GDP and trade balance (via exchange rates). Gross domestic income had certain impacts on the distance to default of businesses. This was also a basic indicator measuring the national economic cycle.

• Journal of the Korean Data and Information Science Society
• /
• 제20권5호
• /
• pp.925-931
• /
• 2009

In this paper we propose a variance function estimation method for replicated data based on averages of squared residuals obtained from estimated mean function by the least squares support vector machine. Newton-Raphson method is used to obtain associated parameter vector for the variance function estimation. Furthermore, the cross validation functions are introduced to select the hyper-parameters which affect the performance of the proposed estimation method. Experimental results are then presented which illustrate the performance of the proposed procedure.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 2007년도 춘계학술대회 논문집 전기기기 및 에너지변환시스템부문
• /
• pp.29-33
• /
• 2007

In this paper, time-varying parameter of passive telemetry RF sensor system is estimated using RLS(Rescursive $\leq$* Square) algorithm. In order to overcome the problems such as power limits and complication that general RF sensor system including IC chip has, the principle of inductive coupling is applied to model sensor system The model parameter is rearranged for applying RLS algorithm based on mathematical model to the derived model using inductive coupling principle. Time variant parameter of rearranged model is estimated using forgetting factor, and in case measured data is contaminated by noise and modelling error, the performance of RLS algorithm characterized by the convergence of squared error sum is verified by simulation.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• 제41권9호
• /
• pp.1051-1059
• /
• 1992

In this paper, two general methods using time-series analysis in the functional separation of the myoelectric signal of human arm movements are developed. Autocorrelation, covariance method and sequential least squares algorithm were used to determine the model parameters and the order of signal model to describe six arm movement patterns` the forearm flexion and extension, the wrist pronation and supination, rotation-in and rotation-out. The confidence interval to classify the functions of arm movement was defined by the mean and standard deviation of total squared error. With the error signals of autoregressive(AR) model, the result showed that the highest success rate was obtained in the case of 4th order, and success rate was decreased with increase of order. Autocorrelation was the method of choice for better success rate. This technique might be applied to biomedical and rehabilitation engineering.

• Journal of Korean Society for Quality Management
• /
• 제24권2호
• /
• pp.65-76
• /
• 1996

In general, most industrial products exhibit bath-tub shaped curve for their failure rate functions. This distributional life model can be obtained by the Piecewise Weibull distribution. The least squares, maximum likelihood, and mixed methods of estimating the parameters of the Piecewise Weibull distribution are compared. The comparison is made by using the empirical mean squared errors of (a) the parameter estimates and (b) the estimated change-points, to summarize the results of 1000 simulated samples of three sizes - each 100, 150 and 200. The results are that the mixed method estimation comes to be the best as the sample sizes increase.

• Transactions of the Korean Society of Automotive Engineers
• /
• 제13권4호
• /
• pp.58-65
• /
• 2005

A method measuring angular speed and estimating angular acceleration of an automotive wind shield wiper pivot with limited resources has been proposed. Limited resources refer to the fact that processes cannot be operated in real-time with a regular notebook running a Microsoft Windows. Also, they refer to the fact that data acquisition cards have only two general purpose counters as many generic cards do. An optical incremental encoder has been employed for measuring angular motion. To measure the angular speed of the pivot, periods for the encoder's output pulses have been measured as the speed is related to the reciprocal of the period. Since only information acquired from one counter channel is the magnitude of the angular speed, sign correction is necessary. Also the information for the exact time when a pivot passes left and right dead points is also missing and the situation is inherent to the hardware setup. To find out the zero-crossing time of the angular speed, a linear interpolation technique has been employed. Lastly, to overcome the imperfection of the mechanical encoders, the angular speed has been curve fitted to a spline. Angular acceleration can be obtained by a differentiation of the angular speed.

• The Korean Journal of Applied Statistics
• /
• 제26권3호
• /
• pp.431-440
• /
• 2013

We consider the problem of estimating the parameters of heavy tailed distributions. In general, maximum likelihood estimation(MLE) is the most preferred method of parameter estimation because it has good properties such as asymptotic consistency, normality and efficiency. However, MLE is not always the best solution because MLE is unstable or does not exist in some cases. This paper proposes another parameter estimation method, non-linear least squares(NLS) and compares its performance to MLE. The NLS estimator is achieved by minimizing sum of squared difference between empirical cumulative distribution function(CDF) and a theoretical distribution function. In this article, we compare the NLS method to MLE using simulated data from heavy tailed distributions. The NLS method is shown to perform better than MLE in Burr distribution when the sample size is small; in addition, it performs well in a Frechet distribution.

• Korean Chemical Engineering Research
• /
• 제45권1호
• /
• pp.32-38
• /
• 2007

A batch Process has a multi-way data structure that consists of batch-time-variable axis, so the statistical modeling of a batch process is a difficult and challenging issue to the process engineers. In this study, We applied a statistical process control technique to the general batch process data, and implemented a fault-detection and Statistical process control system that was able to detect, identify and diagnose the fault. Semiconductor etch process and semi-batch styrene-butadiene rubber process data are used to case study. Before the modeling, we pre-processed the data using the multi-way unfolding technique to decompose the data structure. Multivariate regression techniques like support vector regression and partial least squares were used to identify the relation between the process variables and process condition. Finally, we constructed the root mean squared error chart and variable contribution chart to diagnose the faults.