A Comparison of Weighted Least Square and Quantile Regression for Solving Heteroscedasticity in Simple Linear Regression

Fransiska, Welly; Sigit, Nugroho; Rachmawati, Ramya

doi:10.33369/jsds.v1i1.21011

Cited by 3 publications

(2 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are several requirements in the OLS method, that is, the data shall be normally distributed, no autocorrelation and multicollinearity, and homoscedasticity. 46,47 Without these requirements, the OLS method cannot be adopted as the best linear unbiased estimator. Considering these constraints, the researchers decided to use other methods for comparison, that is, LASSO Linear Regression and Ridge Linear Regression.…”

Section: Methodsmentioning

confidence: 99%

An indoor airflow distribution predictor using machine learning for a real-time healthy building monitoring system in the tropics

Faridah,

Utami,

Wijaya

et al. 2024

Building Services Engineering Research and Technology

View full text Add to dashboard Cite

Indoor air quality is the foundation of a good indoor environment. The COVID-19 pandemic further highlighted the importance of providing real-time airflow distribution information within the Building Environmental Monitoring System (BEMS) to minimize the risk of infectious airborne transmission. This paper discusses the process of developing a predictive model for indoor airflow distribution prediction with indoor and outdoor input parameters using machine learning and its implementation in healthy BEMS for a classroom in the tropical climate region of Yogyakarta, Indonesia. This paper encompassed field measurement and simulation involving outdoor climate conditions and the operational status of the classroom’s windows, Air Conditioning units, and fans. Three machine learning models were constructed using OLS, LASSO, and Ridge methods. Datasets for the modeling were generated from CFD model simulations in IES VE and were assessed for correlation. The mean temperature and velocity differences between the CFD model simulation and measurement results are 0.21°C and 0.083 m/s, respectively. Outdoor climate conditions and the operational status of the classroom’s utilities significantly influence the indoor airflow distribution characteristics. The three models indicate a relatively poor performance, where the classroom had a relatively low sensitivity to input changes. However, the best model performance was achieved using the LASSO method, with average values from post-normalization of [Formula: see text] and Root Mean Square Error (RMSE) of 0.336 and 0.077, respectively. The model was implemented in healthy BEMS on the “Platform for Healthy and Energy Efficient Building Management System.” Practical Application: This research proposed a machine learning model of indoor airflow characteristics of a classroom in Yogyakarta. The proposed model can be adapted to produce monitoring systems that best represent the related conditions. The method can be adopted to develop a relatively simple, low-cost sensor or model to monitor an indoor environment. Future studies may explore the results of the real-world implementation in a case study.

show abstract

Section: Methodsmentioning

confidence: 99%

An indoor airflow distribution predictor using machine learning for a real-time healthy building monitoring system in the tropics

Faridah,

Utami,

Wijaya

et al. 2024

Building Services Engineering Research and Technology

View full text Add to dashboard Cite

show abstract

“…4. If homoscedasticity assumption was violated, then heteroscedasticity occurs, WLS regression can be used instead (Fransiska et al, 2022), which giving weights W. The weight that will be tested are √ ̌ . 5.…”

Section: Methodsmentioning

confidence: 99%

Comparison of Weights in Weighted Least Square Method For Handling Heteroscedasticity on Multiple Regression Model

Virgantari,

Widyastiti,

Ir Seno

2024

IJMSC

View full text Add to dashboard Cite

Regression analysis is the most popular and commonly used to determine causality between two or more variables. In regression analysis there are several assumptions that must be held, so that the property of the best linear unbiased estimator (BLUE) is still guaranteed. In fact, we often found violations of the assumptions. One of them was violations of the homoscedasticity or occurs heteroscedasticity. The impact of heteroscedasticity in the regression model is that the ordinary least square (OLS) estimator no longer has a minimum variance although still linear and unbiased. To handle this, weighted least square (WLS) regression is used instead, which giving weights on the observations. But the problem often encountered is choosing which the best weight in WLS method. This paper aimed to compare and determine the best weight among 1/X, 1/ , 1/Y and 1/s in multiple regression model. Human development index factors data, which were obtained from the Indonesian Central Bureau of Statistics, were used. The results showed that the best weight on human development index data was 1/s. The coefficient of determination was 98.7% indicating that the model was very good.

show abstract

Optimizing Potato Disease Classification Using a Metaheuristics Algorithm for Deep Learning: A Novel Approach for Sustainable Agriculture

El-Kenawy,

Alhussan,

Khafaga

et al. 2024

Potato Res.

View full text Add to dashboard Cite

A Comparison of Weighted Least Square and Quantile Regression for Solving Heteroscedasticity in Simple Linear Regression

Cited by 3 publications

References 2 publications

An indoor airflow distribution predictor using machine learning for a real-time healthy building monitoring system in the tropics

An indoor airflow distribution predictor using machine learning for a real-time healthy building monitoring system in the tropics

Comparison of Weights in Weighted Least Square Method For Handling Heteroscedasticity on Multiple Regression Model

Optimizing Potato Disease Classification Using a Metaheuristics Algorithm for Deep Learning: A Novel Approach for Sustainable Agriculture

Contact Info

Product

Resources

About