Deby Fakhriyana, Deby
Unknown Affiliation

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

PERBANDINGAN MODEL ARCH/GARCH MODEL ARIMA DAN MODEL FUNGSI TRANSFER (Studi Kasus Indeks Harga Saham Gabngan dan Harga Minyak Mentah Dunia Tahun 2013 sampai 2015) Fakhriyana, Deby; Hoyyi, Abdul; Widiharih, Tatik
Jurnal Gaussian Vol 5, No 4 (2016): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (597.137 KB) | DOI: 10.14710/j.gauss.v5i4.14720

Abstract

Indonesian Composite Index is a value that used to measure the combined performance of shares listed in stock market. Price of crude oil is one of the factors that affect Indonesian Composite Index. If the prices of crude oil is increasing, it will be responsed by Indonesian goverment directly with also increasing the fuel prices, that will have an impact on Indonesian Composite Index. ARIMA  and transfer function are methods of modeling time series data and it have assumption that the residual models have to be homogen. To overcome violations of those assumption, this study continue to modelling ARCH/GARCH with ARIMA and transfer function approach. The data used in this study are daily of Indonesian Composite Index and West Texas Intermediate (WTI) crude oil prices data from 2013 to 2015. This study gained two models, the first is ARIMA (1,1,[3]) which variance model of ARCH(1), it’s AIC value is equal to 7707,4287. The second is transfer fuction model (1,0,0) which noise model ARMA(0,[1,3) as well as variance model ARCH(1), it’s AIC value equal to 7689,18984. The best model is the one that has smallest AIC value. From this study can be concluded that the best of ARCH/GARCH model is ARCH/GARCH model with transfer function approach. Keywords : Indonesian Composite Index, crude oil prices, ARIMA, transfer function, ARCH/GARCH
PENENTUAN VALUE AT RISK (VAR) PADA PORTOFOLIO BIVARIAT DENGAN PENDEKATAN COPULA GUMBEL Febriani, Karina; Tarno, Tarno; Fakhriyana, Deby
Jurnal Gaussian Vol 13, No 1 (2024): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/j.gauss.13.1.79-87

Abstract

One way to minimize risk in stock investment is stock portfolio. Value at Risk (VaR) is a calculation method that can be used to estimate the risk of a stock portfolio. VaR can be measured by parametric and non-parametric approaches. Calculation of VaR with Monte Carlo simulation assumes the data is normally distributed. Stock return data generally has high volatility so that the residual variance of the model is not constant (heteroscedasticity) and not normally distributed. The ARIMA-GARCH model can be used to solve heteroscedasticity problems. Copula is a tool used to model the combined distribution of residuals from the ARIMA-GARCH model which does not require normality assumptions. Gumbel's copula is copula that has the best sensitivity to high risk. This study uses stock data of PT Bukit Asam Tbk (PTBA) and PT Chandra Asri Petrochemical Tbk (TPIA) for the period April 1 2020 – December 1 2022. The initial step of this research is model stock returns using the ARIMA-GARCH method and then calculate portfolio VaR using the Gumbel’s copula. The results showed that the best model for PTBA is ARIMA(2,0,2) GARCH(1,1) and for TPIA is ARIMA(1,0,0) GARCH (1,1). At the 95% confidence level, the portfolio risk is 2,41%.
IDENTIFIKASI POLA PERILAKU REMAJA DENGAN PATH ANALYSIS Saadah, Ardiana Alifatus; Fakhriyana, Deby; Hersugondo, Hersugondo
Jurnal Gaussian Vol 12, No 4 (2023): Jurnal Gaussian
Publisher : Department of Statistics, Faculty of Science and Mathematics, Universitas Diponegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/j.gauss.12.4.499-508

Abstract

Globalization has an impact on cultural changes in Indonesia. Apart from positive impacts, globalization also has several negative impacts. The decreasing level of politeness in today's teenagers is part of a cultural change that we cannot ignore. Teenagers in this era are reportedly paying less attention to how to act and behave politely. Politeness is the practical application of good manners and etiquette. To improve polite behavior in teenagers, it is important to know factors that might influence polite behavior. This study used psychological theory developed by Ajzen and Fishbein, the Theory of Reasoned Action. Model in this theory consists of four variables, namely attitude, subjective norm, behavioral intention and behavior. Analytical method used in this research is path analysis. Based on the test results, the attitude variable has an effective influence on increasing the polite behavior variable in teenagers. This is because attitude variable not only influence behavioral variable directly, but also indirectly through behavioral intention variable. Furthermore, the increase in polite behavior is significantly influenced by behavioral intentions. Model combination is able to explain 63.06% of the data diversity, while the rest is explained by other variables and error.
Penerapan Metode Clustering untuk Pemetaan Daerah Rawan Bencana di Kabupaten Bojonegoro Kartini, Alif Yuanita; Fakhriyana, Deby
Journal of Mathematics Education and Science Vol. 7 No. 1 (2024): Journal of Mathematics Education and Science
Publisher : Universitas Nahdlatul Ulama Sunan Giri Bojonegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32665/james.v7i1.1448

Abstract

Saat ini data kejadian bencana yang ada di Badan Penanggulangan Bencana Daerah kabupaten Bojonegoro hanya berupa angka-angka kejadian bencana alam dan belum disajikan dalam bentuk peta. Olehkarena itu diperlukan penelitian untuk melakukan pemetaan daerah rawan bencana di kabupaten Bojonegoro. Dari penelitian ini akan menjadi acuan dalam mengakomodir kegiatan mitigasi bencana di kabupaten Bojonegoro. Pada penelitian ini menggunakan metode clustering yakni metode K-Means, K-Medoids, dan X-Means untuk melakukan evaluasi banyaknya cluster yang optimal dan mengembangkan hasil pengelompokan dalam bentuk peta. Data yang digunakan adalah data jumlah kejadian bencana di kabupaten Bojonegoro tahun 2022 yang meliputi jumlah kejadian akibat banjir, jumlah kejadian akibat cuaca ekstrim, jumlah kejadian akibat kebakaran hutan dan lahan, jumlah kejadian akibat kebakaran rumah, jumlah kejadian akibat kekeringan, jumlah kejadian akibat tanah longsor dan jumlah kejadian lain-lain. Berdasarkan nilai Davies Bouldin Index didapatkan hasil bahwa X-Means merupakan metode terbaik dalam pengelompokan wilayah rawan bencana di kabupaten Bojonegoro. Adapun banyaknya cluster yang terbentuk adalah 4 cluster yakni cluster 0 yang terdiri dari 18 kecamatan, cluster 1 yang terdiri dari 7 kecamatan, cluster 2 yang terdiri dari 1 kecamatan dan cluster 3 yang terdiri dari 2 kecamatan. Berdasarkan karakteristik dari hasil pengelompokan dapat disimpulkan bahwa daerah yang paling rawan terjadi bencana di kabupaten Bojonegoro adalah kecamatan Bojonegoro disusul dengan kecamatan Kedungadem dan Temayang.