1 Matakuliah Tahun : J1186 - Analisis Kuantitatif Bisnis : 2009/2010 METODE PERAMALAN Pertemuan 16 2 Framework • Metode Peramalan Regresi • Aplikasi Model (Model Regresi Sederhana dan Regresi Berganda) • Koefisien Determinasi • Interpretasi Hasil dan Analisis Model Bina Nusantara University 3 d). Linear Trend Line Rumus Umum: Y = a + bx dimana: a = intersep x = periode waktu b = kemiringan Y = ramalan untuk periode b N xy x y a Bina Nusantara University N x 2 ( x) 2 y N b x N 4 Linear Trend Projection Model Yi a bX i b>0 Y a b<0 a X Bina Nusantara University 5 Contoh : Linear Trend Projection Perio d 1(x) 2 3 4 5 x=3 Sales (y) xy 8 11 13 15 19 y=13.2 8 22 39 60 95 xy=224 224 5 3 13.2 2.6 2 55 5 3 Bina Nusantara University b a x2 1 4 9 16 25 x2=55 66 15 2.6 5.4 5 5 6 Lanjutan Period Sales (x) (y) 1 8 2 11 3 13 4 15 5 19 6 M A 10.67 13.00 15.67 MA Err. ES ES Err. TP 11 11 4.33 12 3.0 15.8 -0.8 6.00 13.5 5.5 18.4 0.6 16.25 21.0 TP = Trend Projection: Y = 5.4 + 2.6x Bina Nusantara University TP Err. Small errors! 7 Kesalahan Peramalan Kesalahan Peramalan = ( Dt Ukuran yang digunakan: 1. Mean Absolute Deviation (MAD) Ft ) (D MAD t Ft ) n 2. Mean Squared Error (MSE) MSE 2 ( D F ) t t n Pilih metode peramalan yang menghasilkan MAD atau MSE terkecil Bina Nusantara University 8 Model Regresi Linear Shows linear relationship between dependent & explanatory variables – Example: Sales & advertising (not time) Y-intercept ^ Y Dependent (response) variable Bina Nusantara University i Slope = a b X i Independent (explanatory) variable 9 Linear Regression Model Y Yi = a b Xi Error Error Regression line Y^i = a Observed value Bina Nusantara University b Xi X 10 Interpretasi Koefisien Regresi • Slope (b): – Y changes by b units for each 1 unit increase in X. – If b = +2, then sales (Y) is forecast to increase by 2 for each 1 unit increase in advertising (X). • Y-intercept (a): – Average value of Y when X = 0. – If a = 4, then average sales (Y) is expected to be 4 when advertising (X) is 0. Bina Nusantara University 11 Koefisien Determinasi • Answers: ‘How strong is the linear relationship between the variables?’ • Coefficient of correlation - r – Measures degree of association; ranges from -1 to +1 • Coefficient of determination - r2 – Amount of variation explained by regression equation. • Used to evaluate quality of linear relationship. Bina Nusantara University 12 Koefisien Korelasi r n n n i i i n x i yi x i yi n n n n n x i x i n yi yi i i i i Bina Nusantara University 13 Selecting Forecasting Model Example You’re a marketing analyst for Hasbro Toys. You’ve forecast sales with a linear regression model & exponential smoothing. Which model do you use? Year 1 2 3 4 5 Bina Nusantara University Linear Regression Actual Model Sales Forecast 1 1 2 2 4 0.6 1.3 2.0 2.7 3.4 Exponential Smoothing Forecast (.9) 1.00 1.00 1.00 1.90 1.99 14 Linear Regression Year Y F’cast Error Error2 |Error| 1 1 0.6 0.4 0.16 0.4 2 1 1.3 -0.3 0.09 0.3 3 2 2.0 0.0 0.00 0.0 4 2 2.7 -0.7 0.49 0.7 5 4 3.4 0.6 0.36 0.6 Total i 0.0 1.10 2.0 MSE = Σ Error2 / n = 1.10 / 5 = 0.220 MAD = Σ |Error| / n = 2.0 / 5 = 0.400 MAPE = Σ[|Error|/Actual]/n = 1.2/5 = 0.24 = 24% Bina Nusantara University 15 Model Eksponential Smoothing Year Yi F’cast 1.00 1.00 1.00 1.90 1.99 Error Error2 1 1 0.0 0.00 2 1 0.0 0.00 3 2 1.0 1.00 4 2 0.1 0.01 5 4 2.01 4.04 Total 0.3 5.05 MSE = Σ Error2 / n = 5.05 / 5 = 1.01 |Error| 0.0 0.0 1.0 0.1 2.01 3.11 MAD = Σ |Error| / n = 3.11 / 5 = 0.622 MAPE = Σ[|Error|/Actual]/n = 1.0525/5 = 0.2105 = 21% Bina Nusantara University 16 Mana Yang Terbaik??? Linear Regression : MSE = Σ Error2 / n = 1.10 / 5 = 0.220 MAD = Σ |Error| / n = 2.0 / 5 = 0.400 MAPE = Σ[|Error|/Actual]/n = 1.2/5 = 0.24 = 24% Exponential Smoothing: MSE = Σ Error2 / n = 5.05 / 5 = 1.01 MAD = Σ |Error| / n = 3.11 / 5 = 0.622 MAPE = Σ[|Error|/Actual]/n = 1.0525/5 = 0.2105 = 21% Bina Nusantara University 17
© Copyright 2026 Paperzz