No abstract

Outliers are observations that do not follow the distribution of data patterns and can cause deviations from data analysis, so a method for identifying outliers is needed One method in scanning detection is Minimum Vector Variance which is a robust estimator that uses the minimum Vector Variance (VV) criteria. In this study, the MVV method was used to detect outliers in criminality data in Indonesia in 2013 and data that had been entered out by 5% and 10%. The results showed that the MVV method was more effective than the Mahalanobis distance when detecting outliers in data that had been entered out by 5% and 10%.

Abstract:The stability (or instability) of covariance structure does not ensure the stability (or instability) of correlation structure. However, these stability problems are usually tested using the same approach. The most common and widely used approach is based on likelihood ratio criterion. Unfortunately, its sensitivity is not without limitation, its computational efficiency is low and it is quite cumbersome when the number of variables is large. To overcome these obstacles in monitoring the stability of correlation structure, in this paper we propose a multivariate statistical process control approach by using vector variance as multivariate dispersion measure. First, we investigate the sampling distribution of vector variance of standardized variables and then we construct a control chart. Its sensitivity is very promising, its computational efficiency is high and it is practical even when the number of variables is large. The applicability of the proposed method in monitoring the stability of correlation structure in drive rib production process at Indonesian Aerospace Ltd. will be presented.

Outlier adalah suatu observasi yang polanya tidak mengikuti mayoritas data. Outlier dalam kasus multivariat sangat sulit untuk dideteksi, khususnya ketika dimensi lebih dari 2. Kesulitan ini meningkat ketika data set berukuran besar, yakni jumlah variabel menjadi besar. Metode-metode pendeteksian outlier telah lama berkembang dan beberapa digunakan untuk pelabelan outlier sehingga data dapat dipisahkan antara data yang dicurigai sebagai outlier dan data set pada umumnya. Metode-metode tersebut adalah minimum volume ellipsoid disingkat MVE, minimun covariance determinant disingkat MCD, dan minimum vector variance disingkat MVV. Dari ketiga metode tersebut MVV memiliki waktu perhitungan yang paling cepat. Berdasarkan algoritma MVV, kriteria mengurutkan data menggunakan jarak mahalanobis, maka pada paper ini akan dimodifikasi kriteria pengurutan data dengan menghindari penulisan dalam bentuk invers dari matriks variansi kovariansi. Hasil yang diperoleh adalah metode MVV menjadi lebih cepat dengan menggunakan kriteria baru dengan kecermatan yang sama dengan MVV sebelumnya serta akan diaplikan untuk data real dan data simulasi.

Ordinary kriging is one of the geostatistical techniques used for spatial prediction on a spatially distributed random plane. Ordinary kriging is a linear unbiased estimator which is part of a semivariogram system of equations that minimizes errors of variance in estimating mineral resources. The semivariogram model shows optimal results in the estimation using the least square method, the effective minimization method smoothes the data points against the curve on a semivariogram graph, the least square makes the size error efficient in the semivariogram model and has been proven to be effective in reducing errors in the semivariogram model in the case of laterite nickel deposits. at PT. Vale Indonesia Tbk. Thus, conclusively the prediction of unsampled Ni content results is very accurate. This is indicated by the lowest root mean square error (RMSE) in limonite in the exponential model, saprolite in the spherical model, and bedrock in the gaussian model. The greatest value of Ni content in this study was in the saprolite layer.