Interval : Jurnal Ilmiah Matematika

ANALISIS PENGARUH SUB-KELOMPOK EMPAT, PULAU PAPUA DAN BULAN TERHADAP INDEKS HARGA KONSUMEN MENGGUNAKAN RANCANGAN BUJUR SANGKAR LATIN

2026-07-30T06:48:55+00:00

The purpose of this study is to analyze the differences between sub-groups, between provinces, and between months on the Consumer Price Index (CPI), and identify which factors are significantly different through follow-up tests. The method used is the Latin Longitude Scheme (RBSL) with three factors, namely the Group 4 sub-group as the treatment, six provinces as the first controlling factor, and January to June 2025 as the second controlling factor. The results of the variety analysis showed that the sub-group factor had a significant influence on the CPI with F_Calcul = 4,167 > F_Table= 2.71, as well as the provincial factor was significant, with F_Calcul= 3,687 > F_Table= 2.71, while the month factor did not have a significant effect because F_Calcul = 0.629, F_Table = 2.71. Through Duncan's follow-up test, it was found that the sub-group of Housing and Garden Equipment and Supplies (average = 109.43) was significantly different from several other sub-groups, and the province of Central Papua (average = 110.03) had significant differences from other provinces, so that these two factors were the main contributors to the variation in CPI values on the island of Papua.

Keywords: Consumer Price Index, Duncan Test, Latin Longitude Design (LLD), Variety Analysis

ANALISIS PERBEDAAN PENGARUH KELOMPOK PENGELUARAN TERHADAP INFLASI DI PROVINSI BANTEN MENGGUNAKAN RANCANGAN ACAK KELOMPOK (RAK) DENGAN UJI LANJUT DUNCAN

2026-07-30T06:41:00+00:00

Inflation is an important indicator in describing the economic stability of a region. Differences in consumption patterns and price sensitivity cause inflation fluctuations between expenditure groups. This study aims to analyze the influence of expenditure groups and months on inflation values and identify significant differences between groups and months using Duncan's advanced test. The study used a Randomized Block Design (RBD) with inflation values as the response variable, seven expenditure groups as treatments, and January–June as groups. Data analysis was performed using analysis of variance (ANOVA) and Duncan's test. The ANOVA results showed that expenditure groups significantly influenced inflation with . The month factor also had a significant effect with . The Duncan test results showed that the Food, Beverages, and Tobacco group had the highest average inflation of 1.825 and was significantly different from the other groups. In the month factor, March was significantly different from January and February, while April was significantly different from January. Therefore, expenditure groups and months significantly influenced inflation values.

Keywords: Analysis of Variance (ANOVA), Duncan's Advanced Test, Expenditure Group, Inflation, Randomized Block Design (RBD)

PENERAPAN ALGORITMA K-NEAREST NEIGHBOR DENGAN REDUKSI DIMENSI BERBASIS ANALISIS KOMPONEN UTAMA DALAM MENDETEKSI KANKER PAYUDARA

2026-08-04T03:21:37+00:00

Breast cancer is basically curable if detected early. However, 70% of breast cancer patients were only detected when it was already severe, making it incurable and leading to death. In an effort to overcome this problem, this study aims to classify data from routine health checks to detect breast cancer. The data taken from Coimbra University Hospital, Portugal, is classified into healthy or breast cancer classes using the K-Nearest Neighbor with dimension reduction based on Principal Component Analysis. The variables used are age, BMI, glucose, insulin, HOMA, leptin, adiponectin, resistin, and MCP.1. Results show that the Principal Component Analysis reduced nine attributes down to four with a cumulative variance proportion of 76.15%. The dominating variables in the four principal components in descending order were insulin, leptin, and glucose. The data transformed into the principal component axis was then classified using the K-Nearest Neighbor algorithm. The performance of K-Nearest Neighbor in this paper reaches the highest result when k=5, where the F1 score is 72%. This means that the performance of the model in considered good in finding most positive cases (whether a patient has breast cancer) without making too many mistakes.

Keywords: Breast cancer, classification, K-Nearest Neighbor, F1 score, Principal Component Analysis

OPTIMALISASI PORTOFOLIO SAHAM PERBANKAN DENGAN MODEL INDEKS TUNGGAL SHARPE

2026-08-04T04:10:46+00:00

Investment in banking-sector stocks involves a trade-off between expected return and risk. The classical Markowitz approach requires numerous covariance estimates, making it impractical for large stock universes. This study applies Sharpe's Single Index Model to build an optimal portfolio of eight banking stocks listed on the Indonesia Stock Exchange (BJBR, MEGA, BMRI, BBNI, BBTN, NISP, BDMN, and BRIS) using 15 daily return observations from 8 June to 6 July 2026. The Security Characteristic Line was estimated for each stock to obtain alpha, beta, and residual variance, followed by the Elton-Gruber-Padberg cut-off rate procedure to rank stocks by excess return to beta and select the optimal composition. The results show a cut-off rate of 0.986%, with three stocks (BJBR, BRIS, and MEGA) qualifying for the optimal portfolio, weighted 88.6%, 2.9%, and 8.5%. The optimal portfolio has beta 0.51 and daily standard deviation 1.52%, markedly lower than an equally-weighted naive portfolio of the eight stocks (beta 1.00, standard deviation 2.88%), while achieving a comparable return-to-risk ratio. These findings illustrate the practical benefit of the cut-off rate procedure in selecting an efficient risk-return combination, although the short observation period limits the results to a methodological illustration rather than a definitive investment recommendation.

Keywords: banking stocks, cut-off rate, excess return to beta, optimal portfolio, single index model