Nearest neighbours in least-squares data imputation algorithms with different missing patterns

Wasito, Ito; Mirkin, Boris

doi:10.1016/j.csda.2004.11.009

Cited by 29 publications

(14 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, Table 1 shows the imputation results of the missing values of mass. To evaluate the performance of the proposed imputation method, comparison is made between the proposed method and the mean imputation method in terms of the squared imputation error (IE) (see [22]). …”

Section: Imputation Resultsmentioning

confidence: 99%

Bayesian Markov chain Monte Carlo imputation for the transiting exoplanets with an application in clustering analysis

Teng

Hung

Chao³

2015

Journal of Applied Statistics

View full text Add to dashboard Cite

To impute the missing values of mass in the transiting exoplanet data, this paper uses the Frank copula to combine two Pareto marginal distributions. Next, a Bayesian Markov chain Monte Carlo (MCMC) imputation method is proposed. The proposed Bayesian MCMC imputation method is found to outperform the mean imputation method. Clustering analysis can shed light on the formation and evolution of exoplanets. After imputing the missing values of mass in the transiting exoplanet data using the proposed approach, the similarity-based clustering method (SCM) clustering algorithm is applied to the logarithm of mass and period for this complete data set. The SCM clustering result indicates two clusters. Furthermore, the intracluster Spearman rank-order correlation coefficients ρ s for mass and period in these two clusters are 0.401 and −0.188, respectively, at a significance level of 0.01. This result illustrates that the mass and period correlate in an opposite way between the two different clusters. It implies that the formation and evolution processes of these two clusters are different.

show abstract

Section: Imputation Resultsmentioning

confidence: 99%

Bayesian Markov chain Monte Carlo imputation for the transiting exoplanets with an application in clustering analysis

Teng

Hung

Chao³

2015

Journal of Applied Statistics

View full text Add to dashboard Cite

show abstract

“…The NIPALS algorithm was applied to the NSW Ambulance dataset and the obtained PCA model is used to predict the missing values. (40) Of the total number of items on the SF-36 survey (26,982) a total of 538 (1.9%) were not answered and therefore imputed.…”

Section: Discussionmentioning

confidence: 99%

Fit for duty: The health status of New South Wales Paramedics

MacQuarrie

Robertson²,

Micalos³

et al. 2018

Irish J Para

View full text Add to dashboard Cite

IntroductionParamedics are health care workers who respond to medical emergencies. Paramedics exhibit high rates of injury and illness with markers of poor health. The aims of this study were to explore the self-reported health status of paramedics in New South Wales, Australia, and to compare it with that of the Australian general population and to examine paramedics' attitudes towards exercise. MethodsIn 2015, paramedics employed by NSW Ambulance were invited to complete a web-based survey composed of the Medical Outcomes Survey Short Form 36 (SF-36), measures of attitudes towards exercise and demographic information. Normative comparator data for the Australian general population (BMI and SF-36 scores) were sourced from the Household Income Labour Dynamics in Australia 2015 survey. ResultsOf the approximately 3,300 paramedics invited to participate, 747 completed the survey (507 male, 240 female). Mean age and mean years of service were 41.5 ± 9.5 (SD) and 13.6 ± 9.0 respectively. Male paramedics scored higher than females (p<0.001) in the Vitality domain of the SF-36, and regional paramedics had a higher General Health domain score than metropolitan paramedics (p<0.05). Regional male paramedics had higher BMIs than their metropolitan counterparts (28.04 kg/m2 ± 3.99 vs. 26.81 kg/m2 ± 4.67, p = 0.001). Compared to the Australian population, paramedics scored higher in the Physical Function domain (p<0.001) but lower in summary scores for mental and physical health (p<0.001). Paramedics' BMIs were slightly higher than the general population (27.10 ± 4.30 kg/m2 vs. 26.47 ±5.42, p<0.001). Paramedics reported lack of time, family commitments, and lack of motivation and in regional postings: distance to fitness facilities and shift patterns as barriers to exercise. ConclusionsParamedics scored lower on the SF-36 than the general population, which can indicate a lower health-related quality of life. High BMI and low SF-36 scores may be related to a perceived inability to engage in regular exercise. Increasing BMI can be associated with the development of markers of poor health. Attention is needed to ensure that paramedics are "fit for duty". Ambulance management should foster innovative health promotion programs and paramedics need to recognise and value good health.

show abstract

“…It is easy to show that Cov (0) The structure of eigenvalues of Cov(0) has been investigated by Wasito and Mirkin (2006) who found that, of q nonzero eigenvalues, the maximal one is λ=1+(M-q)q whereas all the other q-1 eigen-values are equal to unity. This provides for really elongated shapes, so that we could check whether this change of the shape indeed affects the clustering results.…”

Section: Data and Cluster Structure Parametersmentioning

confidence: 99%

Intelligent Choice of the Number of Clusters in K-Means Clustering: An Experimental Study with Different Cluster Spreads

2010

Self Cite

View full text Add to dashboard Cite

The issue of determining "the right number of clusters" in K-Means has attracted considerable interest, especially in the recent years. Cluster overlap appears to be a factor most affecting the clustering results. This paper proposes an experimental setting for comparison of different approaches at data generated from Gaussian clusters with the controlled parameters of between-and within-cluster spread to model different cluster overlaps. The setting allows for evaluating the centroid recovery on par with conventional evaluation of the cluster recovery. The subjects of our interest are two versions of the "intelligent" K-Means method, ik-Means, that find the right number of clusters one-by-one extracting "anomalous patterns" from the data. We compare them with seven other methods, including Hartigan's rule, averaged Silhouette width and Gap statistic, under six different between-and within-cluster spreadshape conditions. There are several consistent patterns in the results of our experiments, such as that the right K is reproduced best by Hartigan's rule -but not clusters or their centroids. This leads us to propose an adjusted version of iK-Means, which performs well in the current experiment setting.

show abstract

Nearest neighbours in least-squares data imputation algorithms with different missing patterns

Cited by 29 publications

References 18 publications

Bayesian Markov chain Monte Carlo imputation for the transiting exoplanets with an application in clustering analysis

Bayesian Markov chain Monte Carlo imputation for the transiting exoplanets with an application in clustering analysis

Fit for duty: The health status of New South Wales Paramedics

Intelligent Choice of the Number of Clusters in K-Means Clustering: An Experimental Study with Different Cluster Spreads

Contact Info

Product

Resources

About