Moch. Abdul Mukid
Universitas Diponegoro

Published : 74 Documents
Articles

IMPLEMENTASI MARKOV CHAIN MONTE CARLO PADA PENDUGAAN HYPERPARAMETER REGRESI PROSES GAUSSIAN Mukid, Moch. Abdul; Sugito, Sugito
MEDIA STATISTIKA Vol 4, No 1 (2011): Media Statistika
Publisher : Jurusan Statistika FSM Undip

Show Abstract | Original Source | Check in Google Scholar | Full PDF (720.959 KB)

Abstract

This paper studies the implementation of Markov Chain Monte Carlo on estimating the hyperparameter of Gaussian process. Metropolish-Hasting (MH) algorithm is used to generate the random samples from the posterior distribution that can not be generated by a direct simulation method. This algorithm require only a proposal distribution for generating a candidate point. In this paper uniform distribution is choosen as the proposal distribution.   Keywords: Markov Chain Monte Carlo, Gaussian Process, Metropolis-Hasting Algorithm
PENDUGAAN DATA HILANG DENGAN MENGGUNAKAN DATA AUGMENTATION Nova, Mesra; Mukid, Moch. Abdul
MEDIA STATISTIKA Vol 4, No 2 (2011): Media Statistika
Publisher : Jurusan Statistika FSM Undip

Show Abstract | Original Source | Check in Google Scholar | Full PDF (649.446 KB)

Abstract

Data augmentation is a method for estimating missing data. It is a special case of Gibbs sampling which has two important steps. The first step is imputation or I-step where the missing data is generated based on the conditional distributions for missing data if the observed data are known. The next step is posterior or P-step where the estimation process of parameter values ​​from the complete data is conducted. Imputation and posterior steps on the data augmentation will continue to run until the convergence is reached. The estimate of missing data is obtained through the average of simulated values.   Keywords: Missing Data, Data Augmentation, Imputation Step, Posterior Step
PEMODELAN REGRESI PROSES GAUSSIAN PEMODELAN REGRESI PROSES GAUSSIAN MENGGUNAKAN FUNGSI PERAGAM EKSPONENSIAL KUADRAT Mukid, Moch. Abdul
MEDIA STATISTIKA Vol 3, No 1 (2010): Media Statistika
Publisher : Jurusan Statistika FSM Undip

Show Abstract | Original Source | Check in Google Scholar | Full PDF (345.013 KB)

Abstract

Gaussian Process is a collection of random variables where any finite subset of that has a joint multivariate Gaussian distribution. A Gaussian Process is fully specified by its mean and its covariance function. One of the popular covariance functions is squared exponential that has two hyperparameters. In this paper Gaussian Process is used to made a prediction of  the number of clothes produced by PT. APAC INTI CORPORA based on the number of attending employes, the number of overtime employes, the number of brokendown machines and used materials.     Keywords: Gaussian Process, Covariace Functions, Squared Exponential
SIMULASI STOKASTIK MENGGUNAKAN ALGORITMA GIBBS SAMPLING Anifa, Anifa; Mukid, Moch. Abdul; Rusgiyono, Agus
Jurnal Gaussian Vol 1, No 1 (2012): Wisuda Periode Oktober 2012
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (617.001 KB)

Abstract

One way to get a random sample is using simulation. Simulation can be done directly or indirectly. Markov Chain Monte Carlo (MCMC) is an indirectly simulation method. MCMC method has some algorithms. In this thesis only discussed about Gibbs Sampling algorithm. Gibbs Sampling is introduced by Geman and Geman at 1984. This algorithm can be used if the conditional distribution of the target distribution is known. It has applied on two casses, these are generation of bivariate normal random data and parameters estimation using Bayesian method. The data used in this research are the death of pulmonary tuberculosis in ASEAN in 2007. The results obtained are  and with standard error for  and .
PEMETAAN PERSEPSI MERK LAPTOP DI KALANGAN MAHASISWA MENGGUNAKAN ANALISIS KORESPONDENSI BERGANDA (Studi kasus: Mahasiswa Universitas Diponegoro Semarang) Pangastuti, Anissa; Mukid, Moch. Abdul; Sudarno, Sudarno
Jurnal Gaussian Vol 2, No 3 (2013): Wisuda Periode Agustus 2013
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (381.237 KB)

Abstract

The growth of technology makes producer compete creating sophisticated, modern, and practical tools. One of them is competing creating notebook. Some brands that more develop than other brands in the market are Toshiba, Acer, Asus, HP and Dell. This research studies about positioning one brand against other brands in the market and proximity between all brands that affected by some factors. There are, processors, designation notebook for consumer, features, endorsement and guarantee, endurance notebook against damage, and the distant age of notebook consumption when it has damage in hardware for the first time. Because there are so many factors that affecting perceptual mapping and positioning notebook at the market, hence it need to be analyzed using multiple correspondence analysis. Multiple correspondence analysis is an expansion technique from simple correspondence analysis which is a multivariate technique graphically used for exploration data from a multi-way contingency table. The result of this research makes conclusion that there is a similarity between Acer and HP notebook. This statement be marked with proximity of point Acer and HP. It can be seen from the incision magnitude between both of that brands. There are both of them be used for graphic and designing, have the same complete features and for time of damage for the first time that both of that brands experienced are at age > 3 years
KUALITAS PELAYANAN PADA BANK JAWA TENGAH (Studi Kasus : Bank Jateng Cabang Tembalang) Monica, Yosi Dhyas; Hoyyi, Abdul; Mukid, Moch. Abdul
Jurnal Gaussian Vol 2, No 4 (2013): Wisuda Periode Oktober 2013
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (440.788 KB)

Abstract

Service attendance quality is superiority level which is be expected and control above its superiority level for satisfying consumen´s desire. In this case, there are 5 service quality dimensions. Those are tangible, reliability, responsiveness, assurance, and emphaty. This research study was doing at Bank Jateng, where the respondents are the customer of Bank Jateng. Importance Performance Analysis consist of two components, there are quadrant analysis and discrepancy analysis (gap). Quadran analysis can find out the respond of cusumens against variable which has plotted based on interest and performance level from those variables. While gap analysis is being used for perceiving discrepancy between performance of a variable with the expectation from consumen against its variable. Customer Satisfaction Index (CSI) is used for discovering overall satisfaction level of customers. The T2 hotelling control chart is to know the qualiy controlof two or more related quality characteristics. Result of the research is showing that for quadran analysis, those variables which representing 5 service quality dimensions be located spread in different quadran. For gap analysis, the service perormance of a bank represented by 20 variables who representing 5 service quality dimensions, all of which is still under customers expectation. CSI value aa big as 72,22% which is mean customers satisfaction index is on the satisfaction criteria. On T Hotelling chart is said that the process is not restrained statistically yet because there are 4 points is on the top of control chart
RANCANGAN D-OPTIMAL UNTUK REGRESI POLINOMIAL DERAJAT 3 DENGAN HETEROSKEDASTISITAS Budhianti, Naomi Rahma; Widiharih, Tatik; Mukid, Moch. Abdul
Jurnal Gaussian Vol 2, No 2 (2013): Wisuda Periode April 2013
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (520.015 KB)

Abstract

Suatu model hubungan antara variabel prediktor X dan variabel respon Y, dalam hal ini adalah model regresi polinomial derajat 3 dengan heteroskedastisitas yang mempunyai fungsi bobot .  Permasalahan yang muncul adalah bagaimana memilih titik-titik rancangan X yang akan dicobakan sehingga model menjadi signifikan. Rancangan D-Optimal adalah rancangan dengan kriteria keoptimalan meminimumkan variansi estimator parameter. Jika variansi estimator parameter minimum maka diharapkan parameter dalam model menjadi signifikan sehingga model juga signifikan. Kriteria rancangan D-Optimal didapatkan dengan memaksimumkan determinan matriks informasi atau meminimumkan determinan invers matriks informasi. 
PEMBANGKITAN SAMPEL RANDOM MENGGUNAKAN ALGORITMA METROPOLIS-HASTINGS Irwanti, Lies Kurnia; Mukid, Moch. Abdul; Rahmawati, Rita
Jurnal Gaussian Vol 1, No 1 (2012): Wisuda Periode Oktober 2012
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (351.749 KB)

Abstract

Generating random samples can be done directly and indirectly using simulation techniques. This final project will discuss the process of generating random samples and estimate the parameters using an indirect simulation. Indirect simulation techniques used if the target distribution has a complicated shape and high dimension of density functions. Markov Chain Monte Carlo (MCMC) simulation is a solution to do it. One of the algorithms that is commonly used is Metropolis-Hastings. This algorithm uses the mechanism of acceptance and rejection to generate a sequence of random samples. In the example to be discussed, Metropolis-Hastings algorithm is applied to generate random samples of Beta distribution and also estimate the parameter value of the Poisson distribution using a proposal distribution random-walk Metropolis.
ANALISIS PREFERENSI KONSUMEN TERHADAP PRODUK SUSU BERBASIS ANALISIS CONJOINT MENGGUNAKAN METODE PRESENTASI PAIRWISE-COMPARISON (Studi kasus di Beberapa SMP di Kecamatan Banyumanik Kota Semarang) Resmawati, Trianita; Mukid, Moch. Abdul; Safitri, Diah
Jurnal Gaussian Vol 2, No 4 (2013): Wisuda Periode Oktober 2013
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (408.153 KB)

Abstract

In this study aims to help producer or milk companies to know and understand consumer preferences for attributes combination of milk products specifically for adolescent. The method used in this study is the conjoint analysis using pairwise-comparison as a method of presentation. In this research, the attributes that used are the type of milk, flavor, packaging, and fat content. The result of this reserach shows that the packaging is the most important attribute between the other attributes with a relative importance value of 56.13%. The second most importance attribute is flavor of milk with a relative importance value of 38.55%. Fat content was ranked in the third place with a relative importance value of 4.28%, and the type of milk as the fourth attribute with a relative importance value of 1.05%. In addition, the stimuli is desired by consumers for milk products specifically for adolescent are condensed milk, chocolate, canned, and non fat.
ANALISIS EKUITAS MEREK SEPEDA MOTOR HONDA TERHADAP KEPUTUSAN PEMBELIAN DAN PERILAKU PASCA BELI MENGGUNAKAN STRUCTURAL EQUATION MODELLING (SEM) Putranto, Herwindhito Dwi; Hoyyi, Abdul; Mukid, Moch. Abdul
Jurnal Gaussian Vol 2, No 1 (2013): Wisuda Periode Januari 2013
Publisher : Jurusan Statistika UNDIP

Show Abstract | Original Source | Check in Google Scholar | Full PDF (661.621 KB)

Abstract

Research on the implementation of Structural Equation Modelingto analyze the Honda brand equityon purchase decision and post-purchase behavior is based on the strength of the brand equityas a market leader Honda motorcycles in Indonesia for many years. The problem saddressed in this study is how the relationship between brand equity Honda motorcycle on purchase decision and post purchase behavior of consumers. In this study developed six variables consisting of 4 exogenous variables, namely brand awareness, brand response, the impression of quality and product loyalty, to measure brand equityas well as two endogenous variables, ie, purchase decision and post-purchase behavior. The study involved 200 students of the University of Diponegoro as respondents using purposive sampling technique.Structura lequation modeling research is Behavioral Post Buy=Purchasing Decisions + error. From the Goodness of Fittest results, structural equation modelin this study can be used with a value of 70,237 and the Chi-Square probability AGF I1000 and 0951. Brand awareness of 10.1% influence on purchasing decisions and 10% of the post-purchase behavior and is avariable that gives the effect of CR 1477-value ≤2.58. Responses highest brandin fluenceis equal to 32.7% against 32.4% purchase decision and post-purchase behavior. Thusit was concluded that brand awareness does not affect the purchase decision, while there sponse the brand, the impression of quality and product loyalty influence purchasing decisions. Purchasing decisions also provide a positive influence on post-purchase decisions.
Co-Authors Abdul Hoyyi Agung Waluyo Agus Rusgiyono Alan Prahutama Alvita Rachma Devi Ana Kartikawati Anastasia Arinda, Anastasia Anifa Anifa Anissa Pangastuti Arief Rachman Hakim Bella Cynthia Devi, Bella Cynthia Bramaditya Swarasmaradhana Budi Warsito Bunga Maharani, Bunga Catra Aditya Wisnu Aji Dani Al Mahkya Dedy Douglas Harianja, Dedy Douglas Desy Ratnaningrum, Desy Desy Trishardiyanti Adiningtyas, Desy Trishardiyanti Diah Safitri Dini Anggreani Dwi Ispriyanti Dwi Rahmayani, Dwi Dyan Anggun Krismala Elyas Darmawan Endah Cahyaningrum, Endah Erna Puspitasari Etik Setyowati, Etik Fatma Septy Deviana Fitra Ramdhani, Fitra Hasbi Yasin Herwindhito Dwi Putranto Ilham Muhammad Kiki Febri Azriati Laili Isna Nur Khiqmah, Laili Isna Nur Landong Panahatan Hutahaean Lies Kurnia Irwanti Mardison Purba Marta Widyastuti, Marta Maulida Azkiya, Maulida Mesra Nova Muhammad Mujahid Muhammad Rizki Mustafid Mustafid Naomi Rahma Budhianti Niken Anggraini Dewi Novia Agustina, Novia Noviana Nurhayati Nunung Hanurowati, Nunung Nurhikmah Megawati Restu Sri Rahayu, Restu Restu Sri Rahayu, Restu Sri Rita Rahmawati Riza Adi Priantoro, Riza Adi Shaumal Luqman, Shaumal Sri Wahyuningrum Sudarno Sudarno Sugito Sugito Suparti Suparti Syaifudin Karyadi, Syaifudin Syarah Widyaningtyas, Syarah Tatik Widiharih Trianita Resmawati Triastuti Wuryandari Ungu Siwi Maharunti, Ungu Siwi Wetty Anggun Werti, Wetty Anggun Yosi Dhyas Monica Yuciana Wilandari Yudha Subakti, Yudha Yulia Agnis Sutarno Zulfa Wahyu Mardika, Zulfa Wahyu