An overview of applied robust methods

Yuan, Ke‐Hai; Gomer, Brenna

doi:10.1111/bmsp.12230

Cited by 14 publications

(4 citation statements)

References 117 publications

(192 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the bootstrap method can yield consistent SEs, neither NML nor LS is robust to outliers or influential cases. In particular, samples with heavy tails or data contamination can strongly affect the efficiency of the estimates by the LS and NML methods, and robust procedures are needed to control the effect of poor quality of data (Schamberger et al, 2020; Yuan & Gomer, 2021). Our Monte Carlo results indicate that PLS‐SEM mode B yields the greatest SNR in most of the conditions studied.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

Which method delivers greater signal‐to‐noise ratio: Structural equation modelling or regression analysis with weighted composites?

Yuan

Fang

2022

Brit J Math & Statis

Self Cite

View full text Add to dashboard Cite

Observational data typically contain measurement errors. Covariance‐based structural equation modelling (CB‐SEM) is capable of modelling measurement errors and yields consistent parameter estimates. In contrast, methods of regression analysis using weighted composites as well as a partial least squares approach to SEM facilitate the prediction and diagnosis of individuals/participants. But regression analysis with weighted composites has been known to yield attenuated regression coefficients when predictors contain errors. Contrary to the common belief that CB‐SEM is the preferred method for the analysis of observational data, this article shows that regression analysis via weighted composites yields parameter estimates with much smaller standard errors, and thus corresponds to greater values of the signal‐to‐noise ratio (SNR). In particular, the SNR for the regression coefficient via the least squares (LS) method with equally weighted composites is mathematically greater than that by CB‐SEM if the items for each factor are parallel, even when the SEM model is correctly specified and estimated by an efficient method. Analytical, numerical and empirical results also show that LS regression using weighted composites performs as well as or better than the normal maximum likelihood method for CB‐SEM under many conditions even when the population distribution is multivariate normal. Results also show that the LS regression coefficients become more efficient when considering the sampling errors in the weights of composites than those that are conditional on weights.

show abstract

Section: Conclusion and Discussionmentioning

confidence: 99%

Which method delivers greater signal‐to‐noise ratio: Structural equation modelling or regression analysis with weighted composites?

Yuan

Fang

2022

Brit J Math & Statis

Self Cite

View full text Add to dashboard Cite

show abstract

“…Statistical methods detect outliers or deviating patterns throughout the literature; they include using Mahalanobis distance, which quantifies the distance of data points to the centroid of the data in order to detect outliers (Hong et al, 2020) or person-fit analysis based on item response theory models, which statistically identifies people whose response patterns deviate from the usual pattern defined by a model that captures the majority of respondents (Patton et al, 2019). Moreover, other methods such as transforming the data or downweighting suspect responses can help mitigate the impact of aberrant responses (Yuan & Gomer, 2021; Hong & Cheng, 2019).…”

Section: Aberrant Responsesmentioning

confidence: 99%

Data Exclusion in Policy Survey and Questionnaire Data: Aberrant Responses and Missingness

Hong

Carter

Kim

et al. 2023

Policy Insights from the Behavioral and Brain Sciences

View full text Add to dashboard Cite

Data preprocessing is an integral step prior to analyzing data in psychological science, with implications for its potentially guiding policy. This article reports how psychological researchers address data preprocessing or quality concerns, with a focus on aberrant responses and missing data in self-report measures. 240 articles were sampled from four journals: Psychological Science, Journal of Personality and Social Psychology, Developmental Psychology, and Abnormal Psychology from 2012 to 2018. Nearly half of the studies did not report any missing data treatment (111/240; 46.25%), and if they did, the most common approach was listwise deletion (71/240; 29.6%). Studies that remove data due to missingness removed, on average, 12% of the sample. Likewise, most studies do not report any aberrant responses (194/240; 80%), but if they did, they classified 4% of the sample as suspect. Most studies are either not transparent enough about their data preprocessing steps or may be leveraging suboptimal procedures. Recommendations can improve transparency and data quality.

show abstract

“…One of the regression methods to analyze data contaminated by outliers and provide more flexible results is robust regression [10]. Therefore, this study will apply one of the robust regression methods to the case of inflation in Indonesia.…”

Section: Introductionmentioning

confidence: 99%

Robust regression modelling for inflation factor in the Indonesian economy development

Sofro,

Azzahro,

Khikmah

et al. 2024

E3S Web Conf.

View full text Add to dashboard Cite

Inflation is a general and continuous increase in the prices of goods and services over a while. Inflation is measured by the Consumer Price Index (CPI) indicator to calculate the average price change of a package of goods and services households consume over time. Inflation in Indonesia is not in a precarious condition. However, it also states that inflation in Indonesia could rise. This rise can affect the sustainable development goals (SDGs). Therefore, Indonesia needs to have inflation data scrutinized every year. This study aims to examine and show the factors that affect inflation. Robust regression is an important method to analyze data contaminated by outliers and provide more flexible results. So that it can produce robust regression models and determine what factors significantly affect inflation in Indonesia by the research objectives, this study’s research type is experimental research with literature studies. The variable used is Inflation in Indonesia, which is the dependent variable of the study, and the other four independent variables are export value, interest rate, money supply, and exchange rate. After that, a robust regression model with maximum likelihood type (M) estimation, scale (S) estimation, and least median of squares (LMS) estimation will be obtained, and the best model will be selected. The results show that the least median of squares (LMS) estimation is the best with the acquisition of an Akaike information criteria value of -390.1363. Furthermore, the high and low inflation in 20182022 is influenced by interest rates, money supply and export value.

show abstract

An overview of applied robust methods

Cited by 14 publications

References 117 publications

Which method delivers greater signal‐to‐noise ratio: Structural equation modelling or regression analysis with weighted composites?

Which method delivers greater signal‐to‐noise ratio: Structural equation modelling or regression analysis with weighted composites?

Data Exclusion in Policy Survey and Questionnaire Data: Aberrant Responses and Missingness

Robust regression modelling for inflation factor in the Indonesian economy development

Contact Info

Product

Resources

About