scholarly journals IDENTIFIKASI MODEL SELF-EXCITING THRESHOLD AUTOREGRESSIVE DENGAN SWITCHING TWO REGIME (KASUS PADA DATA EKSPOR AGRIKULTUR DI INDONESIA)

2020 ◽  
Vol 14 (4) ◽  
pp. 511-522
Author(s):  
Husnun Nur Ghiffari Putri Riyansyah ◽  
Dewi Retno Sari Saputro ◽  
Bowo Winarno
Keyword(s):  
Time Series ◽  
Parameter Estimation ◽  
Structural Changes ◽  
Linear Time ◽  
Estimation Method ◽  
Least Square ◽  
Sequential Method ◽  
Agricultural Export ◽  
Threshold Autoregressive ◽  
High Level

A time series model that explain the structural changes associated with data in a certain time period is the Threshold Autoregressive (TAR) model. The basic of the TAR model there are some different usage regimes in autoregressive analysis. One model based on TAR is a self-exciting threshold autoregressive (SETAR) model with the same delay parameters for each regimen. The SETAR model has a linear nature in each regime but being nonlinear if the models of each regime are combined. In addition, this model can improve jump data that cannot be captured by linear time series models. This means that the SETAR model has high-level parameters through an appropriate switching regime that is applied to agricultural export data in Indonesia. The purpose of this reseach is to test the estimated SETAR parameter model and apply it to Indonesian agricultural export data. There are three methods that can be done for estimating of parameter of SETAR model, namely the conditional quadratic sequential method, ordinary least square (OLS) and nonlinear least square (NLS). In this research, the two stage parameter estimation method is used with OLS and the second stage parameter estimation is used to optimisze the parameter values ​​that are not significant in the model. In its application, the SETAR model (2,1,1) was obtained to model agricultural export data in Indonesia and the MAPE value was 25%.

Approximations and boundary conditions for continuous-time threshold autoregressive processes

1994 ◽  
Vol 31 (4) ◽  
pp. 1103-1109 ◽  
Author(s):  
Rob J. Hyndman
Keyword(s):  
Time Series ◽  
Boundary Conditions ◽  
Continuous Time ◽  
Linear Time ◽  
Autoregressive Processes ◽  
Threshold Autoregressive ◽  
The Past ◽  
Time Threshold ◽  
Non Linear ◽  
The Way

Continuous-time threshold autoregressive (CTAR) processes have been developed in the past few years for modelling non-linear time series observed at irregular intervals. Several approximating processes are given here which are useful for simulation and inference. Each of the approximating processes implicitly defines conditions on the thresholds, thus providing greater understanding of the way in which boundary conditions arise.

Parameter Estimation of the Mechanical Behavior of the Artery Using a Nonlinear Least Square Method With a Noise Model

2008 ◽  
Author(s):  
Shahrokh Zeinali ◽  
Jongeun Choi ◽  
Seungik Baek
Keyword(s):  
Parameter Estimation ◽  
Blood Vessel ◽  
Mechanical Behavior ◽  
Constitutive Relation ◽  
Environmental Changes ◽  
Constitutive Relations ◽  
Estimation Method ◽  
Least Square ◽  
Noise Model ◽  
Bias Error

Although it is well known that blood vessels adapt and remodel in response to various biomechanical stimuli, quantifying changes in constitutive relation corresponding to environmental changes is still challenging. Especially, when the dimension of blood vessel is small, the uncertainties in experimental measurements become significant and make it difficult to precisely estimate parameters of constitutive relations for mechanical behavior of the blood vessel. Hence without considering measurement error in displacement, a conventional nonlinear least square (NLS) method results in a biased parameter estimation. In this paper, we propose a new parameter estimation method to eliminate such bias error and provide more accurate estimated parameters for a constitutive relation using a weighted nonlinear least square (WNLS) method with a noise model. We first applied the proposed technique to a set of synthesized data with computer generated white noises and compared the fitting results to those of the NLS method without the noise model. We also applied our method to experimental data sets from mechanical tests of rabbit basilar and mouse carotid arteries and studied parameter sensitivity of the constitutive model.

scholarly journals Unification of Internal Resistance Estimation Methods for Li-Ion Batteries Using Hysteresis-Free Equivalent Circuit Models

Batteries ◽  
2020 ◽  
Vol 6 (2) ◽  
pp. 32
Author(s):  
S M Rakiul Islam ◽  
Sung-Yeul Park ◽  
Balakumar Balasingam
Keyword(s):  
Parameter Estimation ◽  
Equivalent Circuit ◽  
Electrochemical Impedance ◽  
Estimation Method ◽  
Equivalent Circuit Model ◽  
Parameter Extraction ◽  
Internal Resistance ◽  
Least Square ◽  
Estimation Methods ◽  
Li Ion

Internal resistance is one of the important parameters in the Li-Ion battery. This paper identifies it using two different methods: electrochemical impedance spectroscopy (EIS) and parameter estimation based on equivalent circuit model (ECM). Comparing internal resistance, the conventional parameter estimation method yields a different value than EIS. Therefore, a hysteresis-free parameter identification method based on ECM is proposed. The proposed technique separates hysteresis resistance from the effective resistance. It precisely estimated actual internal resistance, which matches the internal resistance obtained from EIS. In addition, state of charge, open circuit voltage, and different internal equivalent circuit components were identified. The least square method was used to identify the parameters based on ECM. A parameter extraction algorithm to interpret impedance spectrum obtained from the EIS. The algorithm is based on the properties of Nyquist plot, phasor algebra, and resonances. Experiments were conducted using a cellphone pouch battery and a cylindrical 18650 battery.

GSTARI-ARCH MODEL AND APPLICATION ON POSITIVE CONFIRMED DATA FOR COVID-19 IN WEST JAVA

2021 ◽  
Vol 14 (2) ◽  
pp. 146-157
Author(s):  
Mutik Alawiyah ◽  
Dianne Amor Kusuma ◽  
Budi Nurani Ruchjana
Keyword(s):  
Time Series ◽  
Parameter Estimation ◽  
Spatial Data ◽  
Estimation Method ◽  
Error Variance ◽  
Space Time ◽  
Time Series Model ◽  
Series Data ◽  
Arch Model ◽  
West Java

Time series model that is commonly used is the Box-Jenkins based time series model. Time series data phenomena based on Box-Jenkins can be combined with spatial data, it is called the space time model One model based on Box-Jenkins model with heterogeneous location characteristics is the Generalized Space Time Autoregressive Integrated (GSTARI) model for a model that assumes data is not stationary or has a trend. This paper discusses the development of the GSTARI model with the assumption that the error variance is not constant which is applied to positive data confirmed by Covid-19 in West Java Province, especially in 4 regencies/cities that have cases in the high category from 6 March 2020 until 31 December 2020. Four regencies/cities are Depok City, Bekasi City, Bekasi Regency, and Karawang Regency. Parameter estimation method for the assumption of non-constant error variance can use Autoregressive Conditional Heteroscedasticity (ARCH) method. GSTARI-ARCH modeling procedure followed three Box-Jenkins stages, namely the identification process, parameter estimation and checking diagnostic. Application of the GSTARI-ARCH Model to Covid-19 positive confirmed data in 4 regencies/cities has a minimum value of RMSE in Bekasi City. The plot of forecast results for the four regencies/cities has a similar pattern to the actual data only applicable for a short time for 1-2 days.

Research on Weighted Iterative Stage Parameter Estimation Algorithm of Time Series Model

2014 ◽  
Vol 687-691 ◽  
pp. 3968-3971
Author(s):  
Wei Shan ◽  
Lei Li ◽  
Qun He
Keyword(s):  
Time Series ◽  
Parameter Estimation ◽  
Time Series Analysis ◽  
Arima Model ◽  
Estimation Algorithm ◽  
Least Square ◽  
Good Convergence ◽  
Series Analysis ◽  
Parameter Estimation Algorithm ◽  
Convergence Performance

Time series analysis has been extensively used in many fields, such as system identification, modeling and data predication, and played an important role in system design, planning and performance analysis. The focus of time series application study is how to improve the accuracy and computation speed of the parameter estimation. Many researchers have carried out system modeling study by applying time series analysis and have gained their research results. The traditional methods such as maximum likelihood estimation, moment estimate and least square estimate which exit the defect of low precision, poor convergence and parameter estimation white noises coupling, are mostly utilized in parameter estimation for model. Taking this as basis the data forecasting and anomaly detection are conducted, which is hard to ensure the system’s stability. Different from the traditional algorithm, this paper proposes a new weighted iterative stage parameter estimation algorithm which avoids the coupling with white noise estimation of ARMA model parameter and improves the accuracy of parameter estimation. In theory, this algorithm tends to provide a good convergence performance. The experimental results based on ARIMA model show that the algorithm can improve the accuracy of parameter estimation and provide a good convergence performance.

A novel parameter estimation method for the Weibull distribution on heavily censored data

2019 ◽  
pp. 1748006X1988764 ◽  
Author(s):  
Renyan Jiang
Keyword(s):  
Parameter Estimation ◽  
Censored Data ◽  
Estimation Method ◽  
Least Square ◽  
Likelihood Method ◽  
Estimation Methods ◽  
Distribution Models ◽  
Data Set ◽  
Hybrid Censoring ◽  
Parameter Estimation Method

It is desired to build the life distribution models of critical components (which are assumed to be non-repairable) of a repairable system as early as possible based on field failure data in order to optimize the operation and maintenance decisions of the components. When the number of the systems under observation is large and the observation duration is relatively short, the samples obtained for modeling are large and heavily censored. For such samples, the classical parameter estimation methods (e.g. maximum likelihood method and least square method) do not provide robust estimates. To address this issue, this article develops a hybrid censoring index to quantitatively describe censoring characteristics of a data set, proposes a novel parameter estimation method based on information extracted from censored observations, and evaluates the accuracy and robustness of the proposed method through a numerical experiment. Its applicable range in terms of the hybrid censoring index is determined through an accuracy analysis. The experiment results show that the proposed approach provides much accurate estimates than the classical methods for heavily censored data. A real-world example is also included.

scholarly journals A New Quantile Estimation Method of Weibull-Rayleigh Distribution

2020 ◽  
pp. 28-37
Author(s):  
A. S. Ogunsanya ◽  
E. E. E. Akarawak ◽  
W. B. Yahya
Keyword(s):  
Parameter Estimation ◽  
Estimation Method ◽  
Simulated Data ◽  
Likelihood Estimation ◽  
Absolute Error ◽  
Rayleigh Distribution ◽  
Least Square ◽  
Estimation Methods ◽  
Data Set ◽  
Total Deviation

In this paper, we compared different Parameter Estimation method of the two parameter Weibull-Rayleigh Distribution (W-RD) namely; Maximum Likelihood Estimation (MLE), Least Square Estimation method (LSE) and three methods of Quartile Estimators. Two of the quartile methods have been applied in literature, while the third method (Q1-M) is introduced in this work. The methods have been applied to simulate data. These methods of estimation were compared using Error, Mean Square Error and Total Deviation (TD) which is also known as Sum Absolute Error Estimate (SAEE). The analytical results show that the performances of all the parameter estimation methods were satisfactory with data set of Weibull-Rayleigh distribution while degree of accuracy is determined by the sample size. The proposed quartile (Q1-M) method has the least Total Deviation and MSE. In addition, the quartile methods perform better than MLE for the simulated data. In particular, the proposed quartile methods (Q1-M) have an added advantage of simplicity in usage than MLE methods.

Sign in / Sign up

Export Citation Format

Share Document