Review of Outlier Detection and Identifying Using Robust Regression Model

Begashaw, Getnet Bogale; Yohannes, Yordanos Berihun

doi:10.11648/j.ijssam.20200501.12

Cited by 14 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Robust regression is used to detect outliers and gives resistant results (Begashaw and Yohannes, 2020). In this study, robust regression requires a weighting function that can minimize the effect of outliers on the model so that the best model will be obtained.…”

Section: Resultsmentioning

confidence: 99%

Earnings quality determinants in pre-corona crisis: another insight from bank core capital categories

Valdiansyah

Murwaningsari

2022

AJAR

View full text Add to dashboard Cite

PurposeDiscretionary accruals are earnings quality proxies that illustrate that the greater the value of discretionary accruals, the greater the practice of earnings management and vice versa. High-quality financial reports (especially earnings quality) are expected to help investors and potential investors to make decisions. This study analyses the factors that affect earnings quality, such as pre-managed earnings, liquidity and efficiency. Furthermore, the authors identify the moderating effect of the governance mechanisms proxied by the proportion of independent commissioners in conventional commercial banks listed on the Indonesia Stock Exchange.Design/methodology/approachThis study uses 226 banking data in the pre-corona crisis period 2013 until 2019. The data were analyzed using EViews 10 for hypothesis and MS Excel for a differential test.FindingsThe results show that pre-managed earnings, liquidity and efficiency affect earnings quality. The governance mechanisms can moderate liquidity and efficiency on earnings quality, while pre-managed earnings cannot be moderated. The different bank categories (BUKU) of earnings management mechanisms are shown for each BUKU (BUKU 1, 3 and 4 perform earnings management by increasing earnings, BUKU 2 lowering earnings). Another thing is information on the earnings quality between BUKU 2 with BUKU 3 and BUKU 4 because of differences in capital and bank operating coverage regulations.Research limitations/implicationsFurther research expects to analyze the factors affecting banking earnings quality concerning applying IFRS 9 (PSAK 71) in Indonesia. Future researchers expect to apply mixed methods to verify the financial statement data and provide comprehensive discussion and genuine insight from their study. Future research requires more samples from companies or an international scale (cross country) to obtain maximum results and be generally accepted.Practical implicationsThis study implies that managers should have more control over pre-managed earnings and bank liquidity as manager's incentive to do earnings smoothing. Managers should also pay attention to cost-efficiency and effective implementation of governance mechanisms to maximize earnings quality. This study also implies that policymakers can encourage commercial banks to apply more prudential principles in terms of a reserve for failed loans to minimize earnings management in banking.Originality/valueThe significance of this study revealed in the discussion of the difference test between bank core capital categories (BUKU) and its relation to earnings quality.

show abstract

Section: Resultsmentioning

confidence: 99%

Earnings quality determinants in pre-corona crisis: another insight from bank core capital categories

Valdiansyah

Murwaningsari

2022

AJAR

View full text Add to dashboard Cite

show abstract

“…An outlier is a data point that differs considerably from other data points. Such an outlier is called influential if the important features of typical analyses would be altered if they were to be retained or deleted from the dataset (Begashaw & Yohannes, 2020). Several different mechanisms can result in outliers in the data, such as sampling errors (Begashaw & Yohannes, 2020).…”

Section: Models For Mean and Quantiles In The Literaturementioning

confidence: 99%

“…Such an outlier is called influential if the important features of typical analyses would be altered if they were to be retained or deleted from the dataset (Begashaw & Yohannes, 2020). Several different mechanisms can result in outliers in the data, such as sampling errors (Begashaw & Yohannes, 2020). In contrast, natural outliers that are not due to a sampling error may also emerge.…”

Section: Models For Mean and Quantiles In The Literaturementioning

confidence: 99%

A robust mixed‐effects parametric quantile regression model for continuous proportions: Quantifying the constraints to vitality in cushion plants

Burger

Merwe

Lesaffre

et al. 2023

Statistica Neerlandica

View full text Add to dashboard Cite

There is no literature on outlier‐robust parametric mixed‐effects quantile regression models for continuous proportion data as an alternative to systematically identifying and eliminating outliers. To fill this gap, we formulate a robust method by extending the recently proposed fixed‐effects quantile regression model based on the heavy‐tailed Johnson‐ distribution for continuous proportion data to the mixed‐effects modeling context, using a Bayesian approach. Our proposed method is motivated by and used to model the extreme quantiles of the vitality of cushion plants to provide insights into the ecology of the system in which the plants are dominant. We conducted a simulation study to assess the new method's performance and robustness to outliers. We show that the new model has good accuracy and confidence interval coverage properties and is remarkably robust to outliers. In contrast, our study demonstrates that the current approach in the literature for modeling hierarchically structured bounded data's quantiles is susceptible to outliers, especially when modeling the extreme quantiles. We conclude that the proposed model is an appropriate robust alternative to the current approach for modeling the quantiles of correlated continuous proportions when outliers are present in the data.

show abstract

“…In the early stages, Zaman and Bulut [8] , [9] and Subzar et al [10] examined how various robust regression methods could be applied to estimate the mean of a finite population based on ratios, especially in the presence of outliers. In a thorough analysis, Begashaw and Yohannes [11] investigated robust regression models for detecting and recognizing outliers. Audu et al [12] employed robust regression techniques to assess how effective exponential-type estimates are for the population mean.…”

Section: Introductionmentioning

confidence: 99%

Dual-type quantile regression approach for mean estimation incorporating sampled and non-sampled data

Alomair,

Iftikhar

2024

Heliyon

View full text Add to dashboard Cite

Review of Outlier Detection and Identifying Using Robust Regression Model

Cited by 14 publications

References 0 publications

Earnings quality determinants in pre-corona crisis: another insight from bank core capital categories

Earnings quality determinants in pre-corona crisis: another insight from bank core capital categories

A robust mixed‐effects parametric quantile regression model for continuous proportions: Quantifying the constraints to vitality in cushion plants

Dual-type quantile regression approach for mean estimation incorporating sampled and non-sampled data

Contact Info

Product

Resources

About