Automatic detection of discordant outliers via the Ueda’s method

Marmolejo‐Ramos, Fernando; Vélez, Jorge Íván Bonilla; Romão, Xavier

doi:10.1186/s40488-015-0031-y

Cited by 13 publications

(17 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to ensure that the observed effects were not due to the type of filter used, a multiverse analysis was carried out (Steegen et al, 2016 ). The reaction time filters compared to the one we chose were: 2 median absolute deviations (MAD) and 2 SD per condition for each participant, and no reaction time filter (see Marmolejo-Ramos et al, 2015 , for a discussion of the most appropriate reaction time filters according to the characteristics of the data; see also Marmolejo-Ramos et al, 2015 ). We did not use a logarithmic transformation, because it is problematic when one condition is significantly slower than the other (Lo and Andrews, 2015 ); judgments of valence were much slower than temporal judgements in the present study.…”

Section: Methodsmentioning

confidence: 99%

When the Sad Past Is Left: The Mental Metaphors Between Time, Valence, and Space

et al. 2018

View full text Add to dashboard Cite

A mental metaphor is a strategy that consists of completing the representation of a concept with structural components of a correlating concept. Three issues were addressed here to deepen our understanding of this mechanism: the use of mental metaphors between abstract concepts, the simultaneous activation of multiple mental metaphors and the importance of the focus of attention on the relevant dimensions of a mental metaphor. In two experiments, participants made temporal or valence judgments (with their left or right hand) on verbs with a negative or positive meaning and conjugated in the past or future form, allowing for the simultaneous activation of the “time is space”, “valence is space,” and “time is valence” mental metaphors. Left-past/right-future and left-negative/right-positive congruency effects were found, and these effects were greater in the temporal and valence judgment tasks, respectively, demonstrating the importance of attentional cuing. Simultaneously, a congruency effect between the abstract concepts of time and valence (past-negative/future-positive) was observed, revealing that a mental metaphor can occur between abstract concepts and that multiple metaphors can be processed simultaneously. These results are discussed in terms of different theories within the field of mental metaphors.

show abstract

Section: Methodsmentioning

confidence: 99%

When the Sad Past Is Left: The Mental Metaphors Between Time, Valence, and Space

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Unlike the CLT, HCCM, and bootstrap approaches, trimming and Winsorizing implicitly assume that the observed data are contaminated by the presence of outliers, which are recognized as extreme cases in the tails of the data distribution (e.g., see Tukey and McLaughlin, 1963 ; Wilcox, 2017 ). Barnett and Lewis ( 1998 ) provide an extensive treatment of outliers; and recent work has focused on their automatic detection (e.g., see Mavridis and Moustaki, 2008 ; Marmolejo-Ramos et al, 2015 ). Stated differently, the use of either trimming or Winsorizing presumes that non-normality is due to the presence of improper data, and these erroneous data are discarded or modified.…”

Section: Addressing Non-normality (And Heteroscedasticity)mentioning

confidence: 99%

How to Address Non-normality: A Taxonomy of Approaches, Reviewed, and Illustrated

2018

View full text Add to dashboard Cite

The linear model often serves as a starting point for applying statistics in psychology. Often, formal training beyond the linear model is limited, creating a potential pedagogical gap because of the pervasiveness of data non-normality. We reviewed 61 recently published undergraduate and graduate textbooks on introductory statistics and the linear model, focusing on their treatment of non-normality. This review identified at least eight distinct methods suggested to address non-normality, which we organize into a new taxonomy according to whether the approach: (a) remains within the linear model, (b) changes the data, and (c) treats normality as informative or as a nuisance. Because textbook coverage of these methods was often cursory, and methodological papers introducing these approaches are usually inaccessible to non-statisticians, this review is designed to be the happy medium. We provide a relatively non-technical review of advanced methods which can address non-normality (and heteroscedasticity), thereby serving a starting point to promote best practice in the application of the linear model. We also present three empirical examples to highlight distinctions between these methods' motivations and results. The paper also reviews the current state of methodological research in addressing non-normality within the linear modeling framework. It is anticipated that our taxonomy will provide a useful overview and starting place for researchers interested in extending their knowledge in approaches developed to address non-normality from the perspective of the linear model.

show abstract

“…Subsequently, an outlier detection method could be applied to the combined data (i.e., uncontaminated data + outlier observations) and the number of outliers detected would be further compared with that initially introduced. In fact, we have recently found that an outlier detection procedure known as the Ueda's method [53] is more likely to detect outliers when the data's distribution becomes more skewed and asymmetric [54]. Using the Ueda's method for the quantification of outliers affecting distributions transformed via the approach proposed herein is a topic for future investigation.…”

Section: Discussionmentioning

confidence: 98%

A new approach to the Box–Cox transformation

Vélez

Correa

Marmolejo‐Ramos

2015

Front. Appl. Math. Stat.

Self Cite

View full text Add to dashboard Cite

We propose a new methodology to estimate λ, the parameter of the Box-Cox transformation, as well as an alternative method to determine plausible values for it. The former is accomplished by defining a grid of values for λ and further perform a normality test on the λ-transformed data. The optimum value of λ, say * λ , is such that the p-value from the normality test is the highest. The set of plausible values is determined using the inverse probability method after plotting the p-values against the values of λ on the grid. Our methodology is illustrated with two real-world data sets. Furthermore, a simulation study suggests that our method improves the symmetry, kurtosis and, hence, the normality of data, making it a feasible alternative to the traditional Box-Cox transformation.

show abstract

Automatic detection of discordant outliers via the Ueda’s method

Cited by 13 publications

References 37 publications

When the Sad Past Is Left: The Mental Metaphors Between Time, Valence, and Space

When the Sad Past Is Left: The Mental Metaphors Between Time, Valence, and Space

How to Address Non-normality: A Taxonomy of Approaches, Reviewed, and Illustrated

A new approach to the Box–Cox transformation

Contact Info

Product

Resources

About