Claim Missing Document
Check
Articles

PEMODELAN TERBAIK DAN PERAMALAN TINGKAT SUKU BUNGA SPN 3 BULAN Mubarak, Fadhlul; Wulandya, Siti Arni; Seran, Karlina; Soleh, Agus Mohamad; Andriansyah, .
Jurnal Kajian Ekonomi dan Keuangan Vol 1, No 3 (2017)
Publisher : Badan kebijakan Fiskal

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31685/kek.v1i3.202

Abstract

Salah satu asumsi dasar ekonomi makro yang masih mengalami kendala dalam pengembangan perangkat analisis model ekonomi yang akurat adalah suku bunga Surat Perbendaharaan Negara (SPN) 3 bulan. Hal ini terutama disebabkan periode data yang tidak teratur karena didasarkan kepada rata-rata yield yang dimenangkan dalam lelang yang dilaksanakan pada periode-periode tertentu. Penelitian ini bertujuan untuk memperoleh model proyeksi tingkat suku bunga SPN 3 bulan dengan memperbandingkan beberapa metode deret waktu yaitu pemulusan spline, pemulusan exponential dan pemulusan moving average, serta pemodelan regresi dengan menggunakan spread dengan yield Surat Utang Negara (SUN) 1 tahun. Hasil dari penelitian ini memperlihatkan bahwa metode yang mendekati kondisi riil adalah metode pemulusan spline dan regresi dengan SUN 1 tahun, dimana pemulusan spline lebih baik untuk proyeksi jangka pendek dan regresi dengan SUN 1 tahun lebih baik untuk proyeksi jangka menengah.
Statistical Downscaling to Predict Monthly Rainfall Using Generalized Linear Model with Gamma Distribution Soleh, Agus M
Informatika Pertanian Vol 24, No 2 (2015): Desember 2015
Publisher : Sekretariat Badan Penelitian dan Pengembangan Pertanian

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (716.358 KB) | DOI: 10.21082/ip.v24n2.2015.p215-222

Abstract

Statistical Downscaling (SDS) models might involve ill-conditioned covariates (large dimension and high correlation/multicollinear). This problem could be solved by a variable selection technique using L1 regularization/LASSO or a dimension reduction approach using principal component analysis (PCA). In this paper, both methods were applied to generalized linear modeling with gamma distribution and compared to predict rainfall models at 11 rain posts in Indramayu. More over, generalized linear model with gamma distribution was used to obtain non-negative rainfall prediction and compared with principal component regression (PCR). Two types of ill-conditioned data with different characteristics (CMIP5 and GPCP version 2.2) were used as covariates in SDS modeling. The results show that three methods (PCR, Gamma-PC, and Gamma-L1) did not demonstrate significant difference in term of Root Mean Square Error (RMSE) after addition of dummy variables (month) in the models. However, a generalized linear modeling with gamma distribution could be considered as the best methods since it provided non-negative rainfall predictions.
PENGENDALIAN KOEFISIEN REGRESI LEAST ABSOLUTE DEVIATION PADA RENTANG BERMAKNA MENGGUNAKAN PROGRAM LINIER Setyono, Setyono; Soleh, Agus Mohamad; Rochman, Nur
Informatika Pertanian Vol 27, No 1 (2018): Juni 2018
Publisher : Sekretariat Badan Penelitian dan Pengembangan Pertanian

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1441.468 KB) | DOI: 10.21082/ip.v27n1.2018.p25-34

Abstract

So far, regression analysis is used to model the mean of response variable as a function of some independent variables, using the least squares (LS) method. In general, the LS method is able to describe well the measure of central tendency, however it is not robust against outliers. Therefore, in certain cases, a regression analysis that minimizes the sum of absolute residuals (least absolute deviation - LAD) is required, which is more robust against outliers. So far, the value of the regression coefficient is not modeled and only depends entirely on the data processed. In some cases, the sign and the value of regression coefficients need to be controlled, in order to be in the meaningful range. The results of this study showed that the modification of the constraints on the LAD regression able to control the regression coefficients to be in the meaningful range. The results of bootstrap showed that distribution of controlled regression coefficients were different from distribution of uncontrolled regression coefficients.
Pengembangan Aplikasi Perangkat Lunak Regresi Komponen Utama Agus Mohamad Soleh
FORUM STATISTIKA DAN KOMPUTASI Vol. 9 No. 2 (2004)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penangangan kasus multikolinier pada analisis regresi linier ganda dilakukan melalui berbagai metode, salah satunya adalah dengan menggunakan regresi komponen utama.  Software-software statistik yang ada saat ini belum memberikan suatu langkah yang mudah dalam melakukan analisis regresi.komponen utama.  Dengan menggunakan bahasa pemrograman C++ dikembangkan software SiRegLin, yang merupakan software untuk pendugaan model linier regresi pada kasus terjadinya multikolinier dengan komponen utama. SiRegLin dikembangkan pada sistem operasi berbasis Microsoft Windows dengan perangkat lunak pengembangan menggunakan Borland C++ Builder versi 6.0. Kata Kunci : Multikolinearitas, Regresi Komponen Utama
MODEL OTENTIKASI KOMPOSISI OBAT BAHAN ALAM BERDASARKAN SPEKTRA INFRAMERAH DAN KOMPONEN UTAMA STUDI KASUS : OBAT BAHAN ALAM/FITOFARMAKA PENURUN TEKANAN DARAH Agus Mohamad Soleh; Latifah K. Darusman; Mohamad Rafi
FORUM STATISTIKA DAN KOMPUTASI Vol. 13 No. 1 (2008)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (97.41 KB)

Abstract

Komposisi kimia yang terkandung dalam ekstrak obat bahan alam merupakan suatu komposisi yang kompleks, dengan demikian pengujian keotentikannya tidak dapat dilakukan melalui pedekatan tunggal.  Salah satu teknik analisis yang dapat menggambarkan secara menyeluruh karakteristik kimia suatu bahan adalah teknik spektroskopi FTIR. Spektra FTIR dihasilkan dari interaksi antara energi sinar inframerah dan komponen kimia penyusun campuran bahan, sehingga suatu spektra FTIR merupakan indentitas khas campuran tersebut. Keotentikan komposisi suatu obat bahan alam pada studi  ini ditentukan berdasarkan pada analisis komponen utama spektra inframerahnya.  Studi dilakukan pada obat bahan alam/fitofarmaka penurun tekanan darah (Tensigard® : terdiri dari ekstrak seledri dan ekstrak daun kumis kucing). Pengukuran spektra inframerah dilakukan terhadap formula obat yang persentase komposisinya ditentukan melalui simplex lattice design. Selain itu pengukuran spektra inframerah juga dilakukan terhadap formula obat dengan mengganti (adulterasi) ekstrak kumis kucing dengan obat sintetis (reserpin) dan ekstrak sambiloto. Berdasarkan plot antara skor komponen utama pertama dan skor komponen utama kedua menunjukkan plot tersebut dapat digunakan untuk mendeteksi komposisi obat, tetapi tidak dapat mendeteksi adanya adulterasi komposisi oleh bahan lain.   Kata Kunci : model otentikasi fitofarmaka, simplex lattice design, komponen utama, tensigard
LASSO : SOLUSI ALTERNATIF SELEKSI PEUBAH DAN PENYUSUTAN KOEFISIEN MODEL REGRESI LINIER Agus Mohamad Soleh; _ Aunuddin
FORUM STATISTIKA DAN KOMPUTASI Vol. 18 No. 1 (2013)
Publisher : FORUM STATISTIKA DAN KOMPUTASI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (621.508 KB)

Abstract

A new method, known as LASSO, has recently developed for selections and shrinkage linear regression methods. The method gives an alternative solution on high correlated data between independent variables, where the least squares produces high variance. Based on simulation this method is not better than forward selection (in the case the parameters contains many zero values) and ridge regression (in the case all parameter values close to zero). Unknowing the true parameter and consistency estimates for all conditions that put the LASSO is better than ridge or forward selection.Keywords : LASSO, least square, forward selection, ridge, cross validation
Pemodelan topik pada dokumen paten terkait pupuk di Indonesia berbasis Latent Dirichlet Allocation Aris Yaman; Bagus Sartono; Agus M. Soleh
Berkala Ilmu Perpustakaan dan Informasi Vol 17 No 2 (2021): December
Publisher : Perpustakaan Universitas Gadjah Mada

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/bip.v17i2.2147

Abstract

Introduction. Fertilizer is one of the most important production factors in the world of agriculture. It is crucial to increase the capacity of technology related to fertilizers. Analysis of patent documents can be one way to analyze technological developments, especially fertilizers. Data Collection Methods. The data used in this research are metadata, especially the title and abstract of a patent document in Indonesia. With the keyword "fertilizer," Patent metadata was processed in the 1945-2017 period. Data Analysis. The LDA model can provide a reasonable interpretation regarding topic modeling based on text data. Results and Discussion. The results find that degree of the patent title is better than the abstract of the patent. The LDA approach can adequately separate the topics of fertilizer patent technology so that it does not have multiple interpretations. Conclusion. Based on the findings, there are nine essential topics in the development of fertilizer technology. There is a phenomenon of the lack of technology collaboration between IPC technology sections.
PEMANFAATAN ENSEMBLE LEARNING DAN PENGINDERAAN JAUH UNTUK PENGKLASIFIKASIAN JENIS LAHAN PADI Arif Handoyo Marsuhandi; Agus Mohamad Soleh; Hari Wijayanto; Dede Dirgahayu Domiri
Seminar Nasional Official Statistics Vol 2019 No 1 (2019): Seminar Nasional Official Statistics 2019
Publisher : Politeknik Statistika STIS

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (294.368 KB) | DOI: 10.34123/semnasoffstat.v2019i1.247

Abstract

Pertanian adalah bidang yang sangat penting di Indonesia, sektor ini di tahun 2017 mencatat penyerapan tenaga kerja sebanyak 29.68% dari total seluruh pekerja (BPS, 2018), namun pentingnya sektor pertanian ini berbanding terbalik dengan data pertanian yang tersedia. Tahun 1998 Badan Pusat Statistik (BPS) bersama Japan International Cooperation Agency (JICA) telah mengisyaratkan overestimasi luas panen sekitar 17,07 persen. Ketidakuratan data pertanian ini kemudian diperbaiki pada tahun 2018 melalui kerjasama para stakeholder dengan menyusun suatu metodologi baru dalam menghitung luas lahan yang diberi nama kerangka sampel area. Selain metodologi yang sudah diperbarui, kemajuan teknologi dan teknik analisis di bidang ilmu pengetahuan juga mendukung perbaikan data pertanian. Citra satelit dan teknik klasifikasi menggunakan ensemble learning dapat dimanfaatkan dalam mengklasifikasikan jenis lahan padi. Pada penelitian ini digunakan citra satelit yang berasal dari United States Geological Survey (USGS) yaitu Landsat 8 dan teknik klasifikasi ensemble learning. Citra satelit dimanfaatkan untuk mengekstrak indeks vegetatif dari koordinat koordinat yang diteliti, sedangkan ensemble learning yang digunakan dalam penelitian ini yaitu Random Forest dan Boosting. Hasil pengolahan data menunjukkan Random Forest memiliki akurasi yang lebih tinggi daripada Boosting yaitu dengan nilai 76,52 untuk Random Forest dan 75,60 untuk Boosting. Keunggulan Random Forest terhadap Boosting tidak hanya dari sisi tingkat akurasi saja namun juga dari kestabilan model yang dibentuk.
A Comparative Study of CatBoost and Double Random Forest for Multi-class Classification Annisarahmi Nur Aini Aldania; Agus Mohamad Soleh; Khairil Anwar Notodiputro
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 7 No 1 (2023): February 2023
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v7i1.4766

Abstract

Multi-class classification has its challenge compared to binary classification. The challenges mainly caused by the interactions between explanatory and responses variable are increasingly complex. Ensemble-based methods such as boosting and random forest (RF) have been proven to handle classification problems. We conducted this research to study multi-class classification using CatBoost, a method developed with gradient boosting and double random forest (DRF), RF’s development that is good to be used when the resulting RF model is underfitting. Analysis was carried out using simulation and empirical data. In the simulation study, we generate data based on the distance between classes: high, medium, and low. The empirical data used is the industrial classification code, namely KBLI. CatBoost and DRF can rightly solve the multi-class classification problem at a high distance, measured by a 100% balanced accuracy score. At a medium distance, CatBoost and DRF produce balanced accuracy scores of 99.25% and 97.54%, respectively, whereas 32.37% and 23.97% at the low distance. In empirical studies, CatBoost’s performance outperforms DRF by 4.27%. All the differences are statistically significant based on the t-test result. We also use LIME to explain individual predictions of CatBoost and learn words that contribute the most to an example class’s prediction.
Penerapan Metode Resampling dan K-Nearest Neighbor dalam Memprediksi Keberhasilan Studi Mahasiswa Program Magister IPB Devi Andrian; Agus Mohamad Soleh; Hari Wijayanto
Xplore: Journal of Statistics Vol. 2 No. 1 (2018): 30 Juni 2018
Publisher : Department of Statistics, IPB

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (653.137 KB) | DOI: 10.29244/xplore.v2i1.79

Abstract

Graduate School IPB (SPs - IPB) has been established for a long time and is believed to produce high quality graduates and highly competitive. However, based on existing data recaps, there are a small number of students who did not graduate, either resigned or Drop Out (DO). It needs to be handled by conducting a selection process for prospective students based on the profile and educational background S1. One of them by applying the method of classification K - Nearest Neighbor (KNN). The response variable used is the success status of the study of prospective students, ie graduated and not graduated. While the explanatory variables used are the profiles and educational background of prospective students. There is an imbalance of data in the data obtained, where the class does not pass much less than the passing class. This can reduce the value of classification accuracy in minority class (sensitivity). So that the handling of data imbalance by using resampling method, either in the form of Random Over Sampling (ROS), Random Under Sampling (RUS), and Random Over-Under Sampling (ROUS). The result of comparison of evaluation result of KNN classification by using k = 1 to 6, resulted in greater sensitivity value when accompanied by the process of handling the data imbalance than without the process of handling the data imbalance, although the accuracy and specificity value becomes smaller. The greatest sensitivity value was obtained when applying the KNN classification method with k = 1, accompanied by the handling of data imbalance by the RUS method, with the mean and median sensitivity values of 0.89 and 0.90, respectively.