Articles

KLASIFIKASI KELOMPOK RUMAH TANGGA DI KABUPATEN BLORA MENGGUNAKAN MULTIVARIATE ADAPTIVE REGRESSION SPLINE (MARS) DAN FUZZY K-NEAREST NEIGHBOR (FK-NN) Kristiani, Yani Puspita; Safitri, Diah; Ispriyanti, Dwi
Jurnal Gaussian Vol 4, No 4 (2015): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (479.943 KB)

Abstract

Good classification method will result on less classification error. Classification method developed rapidly. Two of the existing classification methods are Multivariate Adaptive Regression Spline (MARS) and Fuzzy K-Nearest Neighbor (FK-NN). This research aims to compare the classification of poor household and prosperous household based on per capita income which has been converted according to the poverty line between MARS and FK-NN method. This research used secondary data in the form of result of National Economy and Social Survey (SUSENAS) in Blora subdistrict in 2014. The result of the classification was evaluated using APER. The best classification result using MARS method is by using the combination of BF= 76, MI= 3, MO= 1 because it will result on the smallest Generalized Cross Validation (GCV) and the APER is 10,119 %. The best classification result using FK-NN method is by using K=9 because it will result on the smallest error and the APER is 9,523 %. The APER calculation shows that the classification of household in Blora subdistrict using FK-NN method is better than using MARS method. Keywords: Classification, MARS, FK-NN, APER, SUSENAS, Blora
PERBANDINGAN ANALISIS KLASIFIKASI MENGGUNAKAN METODE K-NEAREST NEIGHBOR (K-NN) DAN MULTIVARIATE ADAPTIVE REGRESSION SPLINE (MARS) PADA DATA AKREDITASI SEKOLAH DASAR NEGERI DI KOTA SEMARANG Merluarini, Bisri; Safitri, Diah; Hoyyi, Abdul
Jurnal Gaussian Vol 3, No 3 (2014): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (325.515 KB)

Abstract

Classification methods have been developed and two of the existing are K-Nearest Neighbor (K-NN) and Multivariate Adaptive Regression Spline (MARS). The purpose of this research is comparing the classification of public elementary school accreditation in Semarang city with K-NN and MARS methods. This research using accreditation data with the result of eight accreditation components in public elementary school that has A accreditation (group 1) and B accreditation (group 2) in Semarang city. To evaluate the classification method used test statistic  Press’s Q, APER, specificity, and sensitivity. The best classification results of the K-NN method is when using K=5 because it produces the smallest error rate and obtained information that the correct classification data are 159 and the misclassification data are 9. The best classification result of the MARS method is when using combination BF=32, MI=2, MO=1 because it produces the smallest Generalized Cross Validation (GCV) and obtained information that the correct classification data are 164 and the misclassification data are 4. Based on analyze result, Press’s Q showed that both methods are good as classification or statistically significant to classify the public elementary school in Semarang city based of the accreditation. APER, specificity, and sensitivity showed that classify of public elementary school accreditation in Semarang city using MARS method is better than K-NN method.
ANALISIS INTERVENSI KENAIKAN HARGA BBM TERHADAP PERMINTAAN BBM BERSUBSIDI PADA SPBU SULTAN AGUNG SEMARANG JAWA TENGAH Ahmad, Fandi; Rahmawati, Rita; Safitri, Diah
Jurnal Gaussian Vol 4, No 1 (2015): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (787.902 KB)

Abstract

Fuel consumption is always interesting to study, in addition to the use of which is used by all the community but also because of the critical role of fuel as an indicator to determine the price of other staples. Not surprisingly, changes in fuel prices polemical definitely interesting to study. In this subject specifically on the impact of the fuel price hike subsidized fuel demand. Changes in fuel price (hike) will have an impact on people's behavior in anticipation of the event. Most people will take the step to buy fuel in bulk prior to the date of determination of the increase in fuel prices, resulting in a surge in demand for fuel. Intervention model is a time series model that can be used to model and predict the data containing the intervention of external factors. In the intervention model, there are two functions, namely the step and pulse functions. Step function is a form of intervention that occurs within a long period of time while the pulse function is a form of intervention that occurs only within a certain time. Based on the analysis suggests that the impact of the use of gasoline and diesel at the pump Sultan Agung Semarang wear both pulse function because the impact was immediate and occur only in a short time                                                                                                                                      Keywords: subsidized BBM, time series, intervention models, pulse function, step function
PENGUKURAN RISIKO KREDIT DAN PENGUKURAN KINERJA DARI PORTOFOLIO OBLIGASI Rizky, Bimbi Ardhana; Sudarno, Sudarno; Safitri, Diah
Jurnal Gaussian Vol 7, No 1 (2018): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (505.269 KB)

Abstract

Except getting coupon as a profit, there is loss probability in bond investment that is credit risks investment. One way to measure the credit risk of a bond is to use the credit metrics method. It uses the ratings of the bond issuer company and the transition rating issued by the rating company for its calculations. Mean Variance Efficient Portfolio (MVEP) can be used to make an optimal portfolio so that risk can be obtained to a minimum. An assessment of portfolio performance is needed  to increase confidence to invest. Sharpe index can measure portfolio performance based on return value of bond. In this case, study has been conduct in two bonds which are Obligasi Berkelanjutan I Bank BTN Tahap II Tahun 2013 and Obligasi Berkelanjutan I PLN Tahap I Tahun 2013 Seri B. The optimum portfolio formed results 67,96% proportion for the first bond and 32,04% for the second bond. For the result, and there is Rp239,4235(billion) of portfolio risk formed. And there is 0,212496for Sharpe index performance assessment portfolio. Keywords: Bond, portfolio, credit risk, credit metrics, Mean Variance Efficient Portfolio, Sharpe index
PEMISAHAN DESA/KELURAHAN DI KABUPATEN SEMARANG MENURUT STATUS DAERAH MENGGUNAKAN ANALISIS DISKRIMINAN KUADRATIK KLASIK DAN DISKRIMINAN KUADRATIK ROBUST Kurniasari, Afianti Sonya; Safitri, Diah; Sudarno, Sudarno
Jurnal Gaussian Vol 3, No 1 (2014): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (397.983 KB)

Abstract

Semarang Regency is one of 29 counties and 6 towns in Central Java province. In the district there are rural areas and urban areas. Discriminant analysis is a technique related to the separation of objects into different groups that have been set previously, thus, discriminant analysis can be used to separate village in Semarang Regency into urban or rural groups. Linear discriminant analysis assumes that the covariance matrix of the two groups are equal, If the assumption of equality covariance matrix is denied, function of quadratic discriminant can be used for classification. Classical estimation for the sample mean vector and sample covariance matrix is very sensitive to the presence of outliers in the observations and the functioning of the separation can be non-robust. Estimators that can be used to cope with data containing outliers are the Minimum Covariance Determinant. Robust discriminant analysis is obtained by replacing the mean and covariance matrix using the classic MCD estimator. After analysis is performed, obtained result the data of 2011 Village Potential contains outlier, so that the robust quadratic discriminant analysis more appropriate because it can provide precision the results of separation 89,79% while classical quadratic discriminant analysis give exactness of 87,23%.
PERBANDINGAN METODE K-MEANS DAN METODE DBSCAN PADA PENGELOMPOKAN RUMAH KOST MAHASISWA DI KELURAHAN TEMBALANG SEMARANG Budiman, Sisca Agustin Diani; Safitri, Diah; Ispriyanti, Dwi
Jurnal Gaussian Vol 5, No 4 (2016): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (478.624 KB)

Abstract

Students as well as community or household, as well as economic activities daily, including consumption. The student needs to choose a place to stay is also one form of consumption activities. There are many factors that affect student preferences in the selection of boarding houses, including price, amenities, location, income, lifestyle, and others. The rental price boarding and facilities offered significant positive effect on student preferences in choosing a boarding house. Based on rental rates and facilities it offered to do the grouping in order to know the condition of the student boarding house in the Village Tembalang. Grouping is one of the main tasks in data mining and have been widely applied in various fields. The method used to classify is K-Means and DBSCAN with a number of groups of three. Furthermore, the results of both methods were compared using the Silhouette index values to determine which method is better to classify the student boarding house. Based on the research that has been conducted found that the K-Means method works better than DBSCAN to classify the student boarding house as evidenced by the value of the Silhouette index on K-Means of 0.463 is higher than the value at DBSCAN Silhouette index is equal to 0.281. Keywords: student boarding houses, data mining, clustering, K-Means, DBSCAN
PERBANDINGAN REGRESI KOMPONEN UTAMA DENGAN REGRESI KUADRAT TERKECIL PARSIAL PADA INDEKS PEMBANGUNAN MANUSIA PROVINSI JAWA TIMUR Sinaga, Vetranella .T.R.A.; Safitri, Diah; Rahmawati, Rita
Jurnal Gaussian Vol 8, No 4 (2019): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

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

Abstract

The multiple regression classic assumptions are used to give linear unbiased and minimum variance estimator. In Human Development Index (HDI) and influencing factors in East Java, there are two variables with VIF more than 10 so the assumption of non-multicollinearity is not fulfilled. Principal component regression (PCR) and partial least squares regression (PLS-R) can solve this problem. By doing principal component analysis, there are two linear combinations to take as the new   independent variables which are free from collinearity. In the PLS-R, NIPALS algorithm is used to calculate the components and other structures and to estimate the parameter. While in PCR all independent variables are significant, the percentage of households with drinking water is feasibles is not significant in the model. PLS-R’s  is 95,85% is greater than PCR’s  = 93,42%. PCR’s PRESS = 81,78 is greater than PLS-R’s PRESS = 61,0595.Keywords: Human Development Index (HDI), Multicollinearity, Principal Component Regression, Partial Least Squares Regression, , PRESS
KLASIFIKASI KEIKUTSERTAAN KELUARGA DALAM PROGRAM KELUARGA BERENCANA (KB) DI KOTA SEMARANG MENGGUNAKAN METODE MARS DAN FK-NNC Hakim, Aryono Rahmad; Safitri, Diah; Sugito, Sugito
Jurnal Gaussian Vol 5, No 3 (2016): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (366.753 KB)

Abstract

Classification method is a statistical method for grouping or classifying data. A good classification method will produce a little bit of misclassification. Classification method has been greatly expanded and two of the existing classification methods are Multivariate Adaptive Regression Spline (MARS) and Fuzzy k-Nearest Neighbor in Every Class (FK-NNC). This study is aimed to compare a classification of Keluarga Berencana  participation based on suspected factors that affect them between the methods of MARS and FK-NNC. This study uses secondary data which one is the participation of Keluarga Berencana in Semarang on 2014. Evaluation of errors use an Apparent Error Rate (APER). In the method MARS best classification results is obtained with the combination of BF = 24, MI = 3, MO = 0 for generating a smallest Generalized Cross Validation (GCV) value and  the APER is obtained by 19%. While FK-NNC method is obtained the best classification results in k = 3 for generating the greatest accuracy of classification value and APER value is obtained by 22%. Based on APER (Apparent Error Rate) calculation, it shown that the classification of family participation in Keluarga Berencana (KB) programs in Semarang using MARS method is better than FK-NNC method.Keywords: Classification, MARS, FK-NNC, APER, Keluarga Berencana
PEMODELAN INDEKS HARGA SAHAM GABUNGAN (IHSG) MENGGUNAKAN MULTIVARIATE ADAPTIVE REGRESSION SPLINES (MARS) Darmawanti, Ndaru Dian; Suparti, Suparti; Safitri, Diah
Jurnal Gaussian Vol 3, No 4 (2014): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (533.391 KB)

Abstract

Composite Stock Price Index (CSPI) is a historical information about the movement of joint-stock until a certain date. CSPI is often used by inventors to see a representation of the overall stock price, it can analyze the possibility of increase or decrease in stock price. Following old examination, some economy macro variables affecting CSPI is inflation, interest rate,and exchange rate the Rupiah againts the u.s.dollar. MARS method is particularly suitable to analyze a CSPI because many variables that affected. Furthermore, in the real world is very difficult to find a spesific data pattern. The analysis is MARS analysis. The purpose is an obtained a MARS model to be used to analyze the CSPI movement’s. Selection MARS model can be used CV method. The MARS model is an obtained from combination of BF, MI, dan MO. In this case, happens the best models with BF=9, MI=2, dan MO=1. Accuracy for MARS model can see MAPE values is 14,32588% it means the model can be used.Keyword: CSPI, economy macro, MARS, CV, MAPE.
PENDUGAAN DATA HILANG PADA RANCANGAN ACAK KELOMPOK LENGKAP DENGAN ANALISIS KOVARIAN Fitri, Vina Riyana; Wuryandari, Triastuti; Safitri, Diah
Jurnal Gaussian Vol 3, No 3 (2014): Jurnal Gaussian
Publisher : Departemen Statistika FSM Undip

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (371.804 KB)

Abstract

Analysis of Covariance (ANCOVA) is mostly used in the analysis of research or experimental design. ANCOVA is the combination between regression analysis and Analysis of Variance (ANOVA). ANCOVA were used because there are some concomitant variable, which is variable that difficult to control by the researchers but an impact on observed the response variable. The purpose from concomitant variable is reduces variability in the experiment. If there is missing data on Randomized Complete Block Design (RCBD) the first must be done estimating the missing data before ANCOVA done. ANCOVA on RCBD with complete data or missing data isn’t much different, if there are missing data, the degrees of freedom is reduced by the total amount of missing data and the sum of square treatment reduced by the value of the bias. Application of tensile strength of the glue experiment to the case ANCOVA on RCBD with one missing data show no effect of treatment and group by the tensile strength of the glue. For Fe toxicity experiment with two missing data are found only treatment effect to Fe texicity. Based on value from the coefficient of variance for one missing data and two missing data showed that ANCOVA is more appropriately used than ANOVA.
Co-Authors Aan Rosiatun Abdul Hoyyi Afianti Sonya Kurniasari Agus Rusgiyono Agustifa Zea Tazliqoh Alan Prahutama Amelia Crystine Anik Nurul Aini, Anik Nurul Annisa Pratiwi, Annisa Arlita, Erna Musri Arsyil Hendra Saputra Artha Ida Sri Anggriyani, Artha Ida Sri Aryono Rahmad Hakim, Aryono Rahmad Aulia Putri Andana, Aulia Putri Bisri Merluarini Budi Warsito Desy Trishardiyanti Adiningtyas, Desy Trishardiyanti Dhinda Amalia Timur Di Asih I Marudani Di Asih I Maruddani Dwi Ispriyanti Esti Pratiwi Evi Yulia Handaningrum Fandi Ahmad Galuh Ayu Prameshti Handini, Juria Ayu Hardi, Desy Tresnowati Hasbi Yasin Hasibuan, Maryam Jamilah An Imam Nur Sholihin Indri Puspitasari, Indri Kishatini Kishartini Lailly Rahmatika, Lailly Maruddani, Di Asih Mekar Sekar Sari Moch. Abdul Mukid Muhamad Faliqul Asbah Muhammad Abid Muhyidin Muhammad Sunu Widianugraha, Muhammad Sunu Mustafid Mustafid Nariswari Diwangkari, Nariswari Natanael, Dimas Kevin Ndaru Dian Darmawanti Nunik Nurhasanah Nur Musrifah Rohmaningsih, Nur Musrifah Nuril Faiz Nurissalma Alivia Putri Octafinnanda Ummu Fairuzdhiya Onny Kartika Hitasari, Onny Kartika Paramita Indrasari Puti Cresti Ekacitta Rahma Nurfiani Pradita, Rahma Nurfiani Rahmawan, Setya Adi Ramadhani, Puput Revaldo Mario, Revaldo Ridha Ramandhani, Ridha Rita Rachmawati, Rita Rita Rahmawati Rizal Yunianto Ghofar Rizka Asri Brilliani, Rizka Asri Rizky, Bimbi Ardhana Rose Debora Julianisa, Rose Debora Sari, Sasmita Kartika Sherly Candraningtyas Sinaga, Vetranella .T.R.A. Sisca Agustin Diani Budiman, Sisca Agustin Diani Sucihatiningsih Dian Wisika Prajanti Sudarno Sudarno Sugito Sugito Suparti Suparti Syilfi Syilfi Tarno Tarno Tatik Widiharih Trianita Resmawati Triastuti Wuryandari Tyas Estiningrum Vierga Dea Margaretha Sinaga, Vierga Dea Margaretha Vina Riyana Fitri Wella Rumaenda, Wella Winastiti, Lugas Putranti Yani Puspita Kristiani, Yani Puspita Yogi Setiyo Pamuji, Yogi Setiyo Yuciana Wilandari Zia, Nabila Ghaida