scholarly journals The Curve Estimation of Combined Truncated Spline and Fourier Series Estimators for Multiresponse Nonparametric Regression

Mathematics ◽  
2021 ◽  
Vol 9 (10) ◽  
pp. 1141
Author(s):  
Helida Nurcahayani ◽  
I Nyoman Budiantara ◽  
Ismaini Zain
Keyword(s):  
Fourier Series ◽  
Nonparametric Regression ◽  
Estimation Method ◽  
Size Variation ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Regression Curve ◽  
Curve Estimation ◽  
Weighted Least Square ◽  
Proposed Model

Nonparametric regression becomes a potential solution if the parametric regression assumption is too restrictive while the regression curve is assumed to be known. In multivariable nonparametric regression, the pattern of each predictor variable’s relationship with the response variable is not always the same; thus, a combined estimator is recommended. In addition, regression modeling sometimes involves more than one response, i.e., multiresponse situations. Therefore, we propose a new estimation method of performing multiresponse nonparametric regression with a combined estimator. The objective is to estimate the regression curve using combined truncated spline and Fourier series estimators for multiresponse nonparametric regression. The regression curve estimation of the proposed model is obtained via two-stage estimation: (1) penalized weighted least square and (2) weighted least square. Simulation data with sample size variation and different error variance were applied, where the best model satisfied the result through a large sample with small variance. Additionally, the application of the regression curve estimation to a real dataset of human development index indicators in East Java Province, Indonesia, showed that the proposed model had better performance than uncombined estimators. Moreover, an adequate coefficient of determination of the best model indicated that the proposed model successfully explained the data variation.

scholarly journals Mixed Spline Smoothing and Kernel Estimator in Biresponse Nonparametric Regression

2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Dyah P. Rahmawati ◽  
I. N. Budiantara ◽  
Dedy D. Prastyo ◽  
Made A. D. Octavanny
Keyword(s):  
Nonparametric Regression ◽  
Kernel Estimator ◽  
Least Square ◽  
Spline Smoothing ◽  
Nonparametric Regression Model ◽  
Weighted Least Square ◽  
Proposed Model ◽  
Mixed Estimator ◽  
Stage Estimation ◽  
Poor Population

Mixed estimators in nonparametric regression have been developed in models with one response. The biresponse cases with different patterns among predictor variables that tend to be mixed estimators are often encountered. Therefore, in this article, we propose a biresponse nonparametric regression model with mixed spline smoothing and kernel estimators. This mixed estimator is suitable for modeling biresponse data with several patterns (response vs. predictors) that tend to change at certain subintervals such as the spline smoothing pattern, and other patterns that tend to be random are commonly modeled using kernel regression. The mixed estimator is obtained through two-stage estimation, i.e., penalized weighted least square (PWLS) and weighted least square (WLS). Furthermore, the proposed biresponse modeling with mixed estimators is validated using simulation data. This estimator is also applied to the percentage of the poor population and human development index data. The results show that the proposed model can be appropriately implemented and gives satisfactory results.

scholarly journals Nonparametric Regression Model for Longitudinal Data with Mixed Truncated Spline and Fourier Series

2020 ◽  
Vol 2020 ◽  
pp. 1-11
Author(s):  
Made Ayu Dwi Octavanny ◽  
I. Nyoman Budiantara ◽  
Heri Kuswanto ◽  
Dyah Putri Rahmawati
Keyword(s):  
Fourier Series ◽  
Longitudinal Data ◽  
Nonparametric Regression ◽  
Simulated Data ◽  
Real Data ◽  
Least Square ◽  
Consistent Finding ◽  
Weighted Least Square ◽  
Stage Estimation

Existing literature in nonparametric regression has established a model that only applies one estimator to all predictors. This study is aimed at developing a mixed truncated spline and Fourier series model in nonparametric regression for longitudinal data. The mixed estimator is obtained by solving the two-stage estimation, consisting of a penalized weighted least square (PWLS) and weighted least square (WLS) optimization. To demonstrate the performance of the proposed method, simulation and real data are provided. The results of the simulated data and case study show a consistent finding.

scholarly journals Estimator Deret Fourier Dalam Regresi Nonparametrik dengan Pembobot Untuk Perencanaan Penjualan Camilan Khas Madura

2018 ◽  
Vol 4 (1) ◽  
pp. 18-23
Author(s):  
Anisatus Sholiha ◽  
Kuzairi Kuzairi ◽  
M. Fariz Fadillah Madianto
Keyword(s):  
Fourier Series ◽  
Nonparametric Regression ◽  
Semiparametric Regression ◽  
Least Square ◽  
Predictor Variables ◽  
Regression Curve ◽  
Parametric Regression ◽  
Determination Coefficient ◽  
Regression Estimators ◽  
The Relationship

The purpose of regression analysis is determining the relationship between response variables to predictor variables. To estimate the regression curve there are three approaches, parametric regression, nonparametric regression, and semiparametric regression. In this study, the estimator form of nonparametric regression curve is analyzed by using the Fourier series approach with sine and cosine bases, sine bases, and cosine bases. Based on Weighted Least Square (WLS) optimization, the estimator result can be applied to model the sale planning of Madura typical snacks. Nonparametric regression estimators with the Fourier series approach are weighted with uniform and variance weight. The best model that be obtained in this study for uniform weight, based on cosine and sine basis with GCV value ​​of 1541.015, MSE value of 0.1375912 and determination coefficient value of 0.4728418%. The best model for variance weight is based on cosine and sine basis with a GCV value of 1541.011, MSE value of 0.1375912 and determination coefficient of 0.4728227%.

scholarly journals Estimation of Heteroskedasticity Semiparametric Regression Curve Using Fourier Series Approach

2020 ◽  
Vol 2 (1) ◽  
pp. 14-20
Author(s):  
Rahmawati Pane ◽  
Sutarman
Keyword(s):  
Fourier Series ◽  
Regression Model ◽  
Variable Parameter ◽  
Predictor Variable ◽  
Semiparametric Regression ◽  
Least Square ◽  
Parameter Vector ◽  
Regression Curve ◽  
Semiparametric Regression Model ◽  
Main Components

A heteroskedastic semiparametric regression model consists of two main components, i.e. parametric component and nonparametric component. The model assumes that any data (x̰ i′ , t i , y i ) follows y i = x̰ i′ β̰+ f(t i ) + σ i ε i , where i = 1,2, … , n , x̰ i′ = (1, x i1 , x i2 , … , x ir ) and t i is the predictor variable. Parameter vector β̰ = (β 1 , β 2 , … , β r ) ′ ∈ ℜ r is unknown and f(t i ) is also unknown and is assumed to be in interval of C[0,π] . Random error ε i is independent on zero mean and varianceσ 2 . Estimation of the heteroskedastic semiparametric regression model was conducted to evaluate the parametric and nonparametric components. The nonparametric component f(t i ) regression was approximated by Fourier series F(t) = bt + 12 α 0 + ∑ α k 𝑐 𝑜𝑠 kt Kk=1 . The estimation was obtained by means of Weighted Penalized Least Square (WPLS): min f∈C(0,π) {n −1 (y̰− Xβ̰−f̰) ′ W −1 (y̰− Xβ̰− f̰) + λ ∫ 2π [f ′′ (t)] 2 dt π0 } . The WPLS solution provided nonparametric component f̰̂ λ (t) = M(λ)y̰ ∗ for a matrix M(λ) and parametric component β̰̂ = [X ′ T(λ)X] −1 X ′ T(λ)y̰

Reproducing Kernel Hilbert space and penalized weighted least square in nonparametric regression

2014 ◽  
Vol 8 ◽  
pp. 7289-7300 ◽  
Author(s):  
Adji Achmad Rinaldo Fernandes ◽  
I Nyoman Budiantara ◽  
Bambang Widjanarko Otok ◽  
Suhartono
Keyword(s):  
Hilbert Space ◽  
Nonparametric Regression ◽  
Reproducing Kernel ◽  
Reproducing Kernel Hilbert Space ◽  
Least Square ◽  
Weighted Least Square

scholarly journals Analyzing The Effect of BI 7-Days Repo Rate on The Jakarta Composite Index Using Nonparametric Regression Approaches Based on Least Square Spline Estimator

2021 ◽  
Vol 17 (3) ◽  
pp. 447-461
Author(s):  
Christopher Andreas ◽  
Feevrinna Yohannes Harianto ◽  
Elfhira Juli Safitri ◽  
Nur Chamidah
Keyword(s):  
Nonparametric Regression ◽  
Composite Index ◽  
Negative Relationship ◽  
Least Square ◽  
Coefficient Of Determination ◽  
Percentage Error ◽  
Absolute Deviation ◽  
Parametric Regression ◽  
Regression Approach ◽  
Repo Rate

During the Covid-19 pandemic, the Indonesia stock market was under great pressure, so that the value of the Jakarta Composite Index (JCI) fluctuated greatly. To maintain economic stability, Bank Indonesia has regulated monetary policy such as setting the BI 7-Days Repo Rate. Analysis of this effect is important to formulate the right policy. This study aims to design the best model in describing the relationship between JCI value and BI 7-Days Repo Rate. The analysis was carried out by using parametric regression approach based on the ordinary least square method and nonparametric regression approach based on least square spline estimator. The results showed that the parametric regression models failed to meet the classical assumptions. Meanwhile, nonparametric regression can produce an optimal model with high accurate prediction, with an overall mean absolute percentage error value of 3.16%. Furthermore, mean square error, coefficient of determination, and mean absolute deviation also show good results. Thus, the effect of the BI 7-Days Repo Rate on the JCI value forms a quadratic pattern, in which a positive relationship is formed when the BI 7-Days Repo Rate is set at more than 4.25% and vice versa for a negative relationship.

scholarly journals Pemodelan Regresi Nonparametrik dengan Estimator Spline Truncated vs Deret Fourier

2021 ◽  
Vol 1 (1) ◽  
pp. 26-36
Author(s):  
Andrea Tri Rian Dani ◽  
Narita Yuri Adrianingsih
Keyword(s):  
Fourier Series ◽  
Least Squares ◽  
Nonparametric Regression ◽  
Case Fatality Rate ◽  
Estimation Method ◽  
Predictor Variable ◽  
Case Fatality ◽  
Ordinary Least Squares ◽  
Regression Modeling ◽  
Fatality Rate

ABSTRAKPendekatan regresi nonparametrik digunakan apabila hubungan antara variabel prediktor dan variabel respon tidak diketahui polanya. Spline truncated dan deret Fourier merupakan estimator dalam pendekatan nonparametrik yang terkenal, karena memiliki fleksibilitas yang tinggi dan mampu menyesuaikan terhadap sifat lokal data secara efektif. Penelitian ini bertujuan untuk mendapatkan estimator model regresi nonparametrik terbaik menggunakan spline truncated dan deret Fourier. Metode estimasi kurva regresi nonparametrik dilakukan dengan menyelesaikan optimasi Ordinary Least Squares (OLS). Kriteria kebaikan model menggunakan GCV, R2 dan MSE. Pemodelan regresi nonparametrik diterapkan pada data Case Fatality Rate (CFR) akibat Demam Berdarah Dengue (DBD) di Indonesia.  Berdasarkan hasil analisis, hasil estimasi dari pemodelan regresi nonparametrik menunjukkan bahwa estimator spline truncated memberikan performa yang lebih baik dibandingkan estimator deret Fourier. Hal ini ditunjukkan dengan nilai R2 dari estimator spline truncated yaitu sebesar 91,80% dan MSE sebesar 0,04, sedangkan dengan estimator deret Fourier diperoleh nilai R2 sebesar 65,44% dan MSE sebesar 0,19.ABSTRACTThe nonparametric regression approach is used when the relationship between the predictor variable and the response variable is unknown. Spline truncated and Fourier series are well-known estimators in the nonparametric approach because they have high flexibility and are able to adjust to the local properties of the data effectively. This study aims to obtain the best nonparametric regression model estimator using the truncated spline and the Fourier series. The nonparametric regression curve estimation method is done by completing the Ordinary Least Squares (OLS) optimization. The criteria for the goodness of the model use GCV, R2, and MSE. Nonparametric regression modeling is applied to Case Fatality Rate (CFR) modeling due to Dengue Hemorrhagic Fever (DBD) in Indonesia. Based on the analysis, the estimation results from the nonparametric regression modeling show that the truncated spline estimator provides better performance than the Fourier series estimator. This is shown by the R2 value of the truncated spline estimator which is 91.80% and the MSE is 0.04, while the Fourier series estimator obtained an R2 value of 65.44% and MSE of 0.19.

Sign in / Sign up

Export Citation Format

Share Document