The Marshall-Olkin Extended Gumbel-weibull Distribution: Properties And Applications

We introduce a new lifetime distribution called Marshall-Olkin extended Gumbel-Weibull. Some properties of distribution such as moments, TL-moments, quantile function, en- tropy, and order statistics are studied. The fexibility of the distribution to model unimodal, monotone shapes as well as unimodal, bimodal, monotone failure rates are presented. The estimators of the parameters of the distribution were obtained using the maximum likeli- hood estimation method. The performance of the maximum likelihood estimates of the Marshall-Olkin extended Gumbel-Weibulll parameters was observed through simulation studies. Two real life applications to illustrate the potentials of the new distribution are presented, and comparison with other distribution having the same baseline is done using goodness-of-test statistics.

Alternative item selection strategies for improving test security in computerized adaptive testing of the algorithm

Salah satu metode estimasi kemampuan yang banyak diaplikasikan pada algoritma Computerized Adaptive Testing (CAT) adalah Maximum Likeli-hood Estimation (MLE). Metode MLE mempunyai kekurangan yaitu ketidakmampuan menemukan solusi estimasi kemampuan peserta tes ketika skor peserta masih belum berpola. Bila ada peserta tes yang memperoleh skor 0 atau skor sempurna, maka untuk menentukan estimasi kemampuan peserta tes umumnya menggunakan model step size. Namun, model step-size tersebut mengakibatkan item exposure. Item exposure merupakan fenomena dimana butir-butir soal tertentu akan lebih sering muncul dibandingkan dengan butir-butir soal yang lain. Hal tersebut membuat tes menjadi tidak aman karena butir-butir soal yang sering muncul akan lebih mudah pula untuk dikenali. Kajian ini mencoba memberikan alternatif strategi dengan cara memodifikasi model step-size dan dilanjutkan dengan merandom hasil perhitungan fungsi informasi yang diperoleh. Berdasarkan hasil kajian didapatkan bahwa alternatif strategi pemilihan butir soal ini mampu menghasilkan kemunculan butir soal yang lebih bervariasi sehingga dapat meningkatkan keamanan tes pada CAT.Kata kunci: item exposure, step-size, adaptive testing AbstractOne method of capability estimation that is widely applied to the Computerized Adaptive Testing (CAT) algorithm is Maximum Likeli-hood Estimation (MLE). The MLE method has the disadvantage of being unable to find a solution to the test taker's ability when the participant's score is not patterned. If there are test takers who get a score of 0 or perfect score, then to determine the ability of the test takers to generally use the step size model. However, the step-size model results in exposure items. The exposure item is a phenomenon where certain items will appear more often than other items. This makes the test insecure because items that often appear will be easier to recognize. This study tries to provide an alternative strategy by modifying the step-size model and proceed by randomizing the results of the calculation of the information function obtained. Based on the results of the study, it was found that alternative item selection strategies were able to produce the appearance of more varied items so as to improve the safety of tests on the CAT.

A new species of the giant pill-millipede genus Sphaerobelum Verhoeff, 1924 from northern Thailand, with an extensive description and molecular characters (Diplopoda: Sphaerotheriida: Zephroniidae)

As a first step towards an inventory of the giant pill-millipedes in Thailand, a new species of the genus SphaerobelumVerhoeff, 1924, S. truncatum n. sp. is described from Nan Province, northern Thailand. A determination key is presentedfor all five known Sphaerobelum species. Clear morphological differences between S. truncatum n. sp. and the other fourSphaerobelum species were found on the anterior telopods. For the first time in Sphaerobelum, the partial mitochondrialCOI gene was sequenced for S. truncatum n. sp. and compared with distance, maximum parsimony and maximum likeli-hood methods to those of species from other giant pill-millipede genera. Sphaerobelum truncatum n. sp. was found to dif-fer from all other analyzed giant pill-millipedes, including species of Zephronia Gray, 1832, by 22–30%, includingnumerous amino acid changes, supporting the separate status of Sphaerobelum among other giant pill-millipede genera.Maximum likelihood and parsimony analyses support the placement of Sphaerobelum in the Zephroniidae. Figures of allrelevant structures of Sphaerobelum truncatum n. sp. are provided to allow the use of these characters in future descriptions of species of the family Zephroniidae.