FORUM STATISTIKA DAN KOMPUTASI
Not a member yet
119 research outputs found
Sort by
MODEL VEKTOR AUTOREGRESSIVE UNTUK PERAMALAN CURAH HUJAN DI INDRAMAYU (Vector Autoregressive Model for Forecast Rainfall In Indramayu )
Get PDFThere are three regions of rainfall that has been formed, each rainfall regions has a variety of homogeneous and there is a correlation between rainfall stations. In each region can be determined rainfall prediction model simultaneously. The model is a model of Vector Autoregressive (VAR) which is an extension of the autoregressive model (AR). Based on this research, we can determine the VAR model by lag 1 or VAR (1) for each region. Region 1 (Anjatan and Sumurwatu), region 2 (Salamdarma and Gantar) and region 3 (Kedokan Bunder and Sudimampir), each of which has a Root Mean Square Error Prediction (RMSEP) of 3.93; 5:03; 4:48; 5.3; 2:18 and 3:53. Correlation value of observations with predictions of rainfall respectively, 0.71; 0.62; 0:57; 0:59; 0.89, and 0.91. Keywords: AR, VAR, RMSEP, correlatio
PREFERENSI MAHASISWA IPB TERHADAP MATA KULIAH METODE STATISTIKA MENGGUNAKAN ANALISIS KONJOIN
Get PDFStatistical method is one of the interdepth courses in Bogor Agricultural University (BAU) therefore, it is necessary to conduct an evaluation in order to know the student\u27s preference towards Statistics Methods course. Conjoint analysis is an analysis that can be used to determine the preference of students on teaching methods of Statistical Methods course. The combination of teaching methods are made using fractional factorial in which the level of factor determined was based on preliminary survey. Sampling techniques that has been used was multistage sampling of students who had took the Statistical Methods course in 2009/2010. Based on conjoint analysis, the module, the number of students, and the time period of lectures are the top three choices. The students tend to prefer materials that are appropriate with their major, modules that are well structured, a communicative lecturer, students as a teacher in review session, the number of student which is less than 50 students per class, and the time period of lecture is between 7-12 am. Keywords : statistical methods, preferences, conjoint analysis
PENERAPAN METODE RANDOM FOREST DALAM DRIVER ANALYSIS
Get PDFDriver analysis is one approach to know which the greatest expalanatory variables influence the response variable. This analysis is well known in marketing research. In this area, explanatatory variables (X) and response variable (Y) ussually are measured by ordinal data and the relationship between those variables is non linier. One of the approach to build model on that situation is random forest. Two important things in random forest are size of random forest and sample size of X. In this research, we worked with simulation to know the size of random forest which give higher accuration and more stabil. The simulation showed that the best condition achieved when the size of random forest is 500 and the sample size of X is 4. Key words : driver analysis, random forest, variable importance
REGRESI KUADRAT TERKECIL PARSIAL MULTI RESPON UNTUK STATISTICAL DOWNSCALING (Multi Response Partial Least Square for Statistical Downscaling)
Get PDFIn climatology partial least square regression (PLSR) can be used as an alternative technique in statistical downscaling based on global circulation model (GCM) output. PLSR is the technique to forecast not only one response but also multi responses to accommodate the correlation among responses. PLSR is compared to PCR (Principal Component Regression). The results show that PLSR is better than PCR and can be used to forecast rainfall simultaneously in more than one rainfall stations relatively as well as in one station. Keywords: statistical downscaling, PLSR, PCR, multi response
MODELLING INGREDIENT OF JAMU TO PREDICT ITS EFFICACY
Get PDFJamu is an Indonesian herbal medicine made from a mixture of several plants. Nowadays, many jamu are produced commercially by many industries in Indonesia. Each producer may have their own jamu formula. However, one is certain; the efficacy of jamu is determined by the composition of the plants used. Thus, it is interesting to model the ingredient of jamu which consist of plants and use it to predict efficacy of jamu. In this analysis, Partial Least Squares Discriminant Analysis (PLSDA) is used in modeling jamu ingredients to predict the efficacy. It is obtained that utilizing the prediction of y ij obtained from PLSDA directly rather than use it to calculate probability of jamu i belong to efficacy j and then use the probability to predict efficacy produces lower False Positive Rate (FPR) in predicting efficacy group. Keywords: Jamu, PLSD
METODE POHON GABUNGAN: SOLUSI PILIHAN UNTUK MENGATASI KELEMAHAN POHON REGRESI DAN KLASIFIKASI TUNGGAL
Get PDFClassification and regression tree has been a widely used tool in various applied fields due to its capability to give excellent predictive analysis. Later on, ensemble tree appeared to enhance simple tree analysis and deals with some of the weakness found in simple techniques. The ensemble tree basically combines predictions values of many simple trees into a single prediction value. This paper is intended as an introductory article to give a brief overview of the available ensemble tree methods which might be found in detail in a variety of reading materials
Analisis Data Longitudinal dengan Metode Regresi Berstruktur Pohon (Kasus Penyakit Kencing Manis)
No full textThis research aimed to analyzed longitudinal data after a tree structure regression method being applied to the data, to group some objects with the same response profile. The comparison of mean profile of all groups is also shown, as well as the comparison of each group\u27s data with ungrouped data. The analyzed longitudinal response data are the glucose content of diabetes patients who cured in M. Jamil Hospital, Padang. Explanatory variables which assumed as the ones those have contribution to the response value are age, sex, relative body weight, kind of diabetes mellitus, complication, the recorded length of symptomps appearance and calorie content of patient\u27s diet. The best tree of glucose content has tix terminal nodes, so that based on the glucose content profile, diabetes patient can be classified into six groups. The classification is based on the variables of kind of diabetes mellitus, age, complication, and relative body weight. The comparison applied to confidence band of glucose content mean shows that the groups have different mean glucose content. Futhermore, it is obvious that the grouped and ungrouped data have different mean of glucose content profile. It is also shown that patients who have recognized diabetes without complication and with neuropati perifer have possibility of increase of glucose content during the curing period. For other group, the treatment given gives results as expected
Perluasan Kueri Menggunakan Peluang Bersyarat (Query Expansion Using Conditional Probability)
No full textUkuran kesamaan merupakan nilai yang digunakan untuk memberikan peringkat kesesuaian dokumen berdasarkan kueri yang diberikan pengguna. Berbagai pengolahan terhadap ukuran kesamaan dilakukan agar semakin banyak dokumen relevan dapat terambil, salah satunya adalah menggunakan peluang bersyarat. Agar dokumen yang ditemukembalikan mempunyai relevansi yang tinggi, maka dapat juga dilakukan perluasan kueri dengan menambahkan istilah yang benar-benar mempunyai keterkaitan tinggi dengan istilah dalam kueri aslinya.Berdasarkan penelitian dan uji coba yang telah dilakukan terhadap sekelompok dokumen, diperoleh kesimpulan bahwa perluasan kueri berbasis peluang bersyarat mampu meningkatkan efisiensi dan nilai presisi dokumen temu kembali. Semakin tinggi peluang kemunculan istilah yang digunakan untuk memperluas kueri terhadap kemunculan istilah lainnya mengakibatkan tingkat relevansi dokumen yang dihasilkan juga semakin tinggi
Penelitian Laboratorium Untuk Mengkaji Beberapa Faktor Yang Mempengaruhi Hasil Voting
No full textPengkajian masalah voting langsung di dunia yang sesungguhnya seperti di MPR/DPR tidak memungkinkan untuk dilakukan. Dengan penelitian laboratorium yang memanfaatkan Teori Induced-value, dapat dikaji faktor-faktor yang mempengaruhi hasil voting sesuai dengan tujuan penelitian ini.Penelitian ini berkesimpulan bahwa dengan voting tertutup jumlah peserta voting yang memilih diluar koalisi lebih besar daripada dengan voting terbuka. Demikian pula dengan imbalan tambahan (polotik uang), semakin besar nilainya semakin besar pula peserta voting memilih diluar koalisi. Tahapan voting dapat juga menghambat jumlah peserta voting memilih diluar koalisi.Kata kunci : penelitian laboratorium, teori induced valu
Pembandingan Hasil Uji F, Uji Quade Dan Uji Friedman Terhadap Pengamatan Hasil Uji Organoleptik (Analisis Statistis Baku Sebagai Suatu Mitos)
No full textUntuk keperluan praktis, beberapa bentuk analisis statistis sering diperlakukan sebagai suatu analisis baku. Dengan analisis baku tersebut hasil yang diperoleh seringkali diterima tanpa memeriksa hal lain yang mungkin memberikan gambaran yang berlainan. Permasalahan yang disoroti dalam artikel ini adalah masalah tranformasi dan uji non-parametrik untuk hasil pengukuran organoleptik. Berdasarkan kasus ini ternyata bahwa, transformasi tidak selalu membuat hasil analisis menjadi lebih baik. Transformasi data tidak selalu meningkatkan kesensitifan uji dan keterandalan analisis, seperti yang selama ini diyakini. Sementara itu, hasil uji non-parametrik memberikan gambaran yang tidak berbeda dengan hasil uji F untuk data yang telah ditransformasi. Dengan membandingkan nilai-p dari hasil dua uji nonparametrik, Friedman dan Quade, uji mana diantara keduanya yang lebih baik belum dapat disimpulkan. Tulisan ini memberikan caution agar peneliti berhati-hati dalam menerapkan analisis ststistis tertentu