Pengembangan Model Regresi Linier Untuk Prediksi Jumlah Penerimaan Mahasiswa Baru (Studi Kasus: Mahasiswa Jurusan Matematika FMIPA UNM)

The number of students from 2000 to 2023 in mathematical courses each year increases and in mathematical education programs decreases. The data obtained is further used to predict the number of students in the next 5 years. The purpose of the study to determine the parameters of simple linear regression models used to predict the number of students on mathematical and mathematical education programs in the next 5 years. The study begins with data retrieval, data review, parameters estimation, regression assumption testing, hypothesis testing, making linear regression models simple, performing model predictions and evaluation. The results of this study showed that the number of students of mathematical programmes had increased, linear regression model y = 39,783 + 0.3774x while in mathematical education prodes had decreased, linear regression model y = 154.29 – 0.7865x. The results of the predicted number of students in 2024 - 2028 will come on mathematical prodes 49,2., 49,6., 50., 50,4., 50,7. At Mathematical Education Prodi 134.6., 133.8., 133.1., 132.3., 131.5. With model test scores on mathematical prodies are MSE=199.4., RMSE=14.12., MAPE=0.236 and on mathematical education prodies are MSE=2.076.44., RMSE=45,568., and MAPE=0.229. Which means that the model is very good and worth using in prediction.