In class exercise- Dry pasta and semester frcst answered

Static Forecasting Step 1 Step 2 Step 4 Step 5 Period Year Semester Enrollment history Deseasonalized Linear Regression Seasonal factor 1 1 Fall 30 37.50 0.80 2 1 Spring 80 41.7 41.33 1.94 3 1 Summer 15 45.0 45.17 0.33 4 2 Fall 40 48.3 49.00 0.82 5 2 Spring 90 53.3 52.83 1.70 6 2 Summer 30 56.67 0.53 7 3 Fall 49 =Step 7 Answer 60.50 Step 3 -Linear regression Slope 3.8 Intercept 33.7 Step 6 Semester Average Fall 0.81 Spring 1.82 Summer 0.43 Step 1 Number the periods Step 2 Find the average of the 3 periods (semesters) Step 3 Run the linear regression to calculate the intercept and slope of data Step 4 Calculate the deseasonalized values for each period Step 5 Calculate the seasonal factors Step 6 Average the seasonal factors Step 7 Multiple the seasonal factors by the linear regression for the forecasted period Notice there is no forecast error in static forecasting. Fall Spring Summer Fall Spring Summer Fall 1 1 1 2 2 2 3 0 10 20 30 40 50 60 70 80 90 100 With Year 3 Fall Forecast

Exercise 7-8 Dry pasta demand MOVING AVERAGE Period Demand Level Forecast Bias 1 517 2 510 3 557 4 498 5 498 516 6 444 501 516 72 72 72 7 526 505 501 -25 25 47 8 441 481 505 64 64 111 9 541 490 481 -60 60 51 10 445 479 490 45 45 96 479 Estimate of standard deviation of fore EXPONENTIAL SMOOTHING Alpha = 0.2 Period Demand Level Forecast Bias 0 498 1 517 502 498 -19 19 -19 2 510 503 502 -8 8 -28 3 557 514 503 -54 54 -81 4 498 511 514 16 16 -65 5 498 508 511 13 13 -53 6 444 495 508 64 64 12 7 526 502 495 -31 31 -19 8 441 489 502 61 61 41 9 541 500 489 -52 52 -10 10 445 489 500 55 55 45 E t A t E t A t Calculate Errors Calculate Errors Estimate demand for the next four weeks using a five- the MAD, MAPE,MSE, bias, and TS in each case. Which

ERRORS MSE MAD MAPE TS 5,184 72 16.2 16.2 1.00 2,895 48 4.7 10.4 0.98 3,278 53 14.4 11.8 2.08 3,347 55 11.0 11.6 0.94 3,082 53 10.1 11.3 1.82 ecast error: 66 ERRORS MSE MAD MAPE TS 372 19 3.7 3.7 -1.00 222 14 1.7 2.7 -2.00 1,111 27 9.7 5.0 -3.00 897 24 3.2 4.6 -2.69 751 22 2.6 4.2 -2.39 1,313 29 14.5 5.9 0.40 1,259 29 5.8 5.9 -0.65 1,560 33 13.7 6.9 1.25 1,682 35 9.5 7.2 -0.29 1,813 37 12.3 7.7 1.20 Percent Error Percent Error 180 280 380 480 580 680 EXPONEN Actual Dem Unit Demand 1 2 3 4 5 80 180 280 380 480 580 680 Moving Average A Unit Demand Graph Using Moving Average Graph Using Exponential Smoothing -week moving average, as well as simple exponential smoothing with α = 0.2. Evaluate h of the two methods do you prefer? Why?

ecast error: 46 lower MAD and MAPE value 1 2 3 4 5 80

6 7 8 9 10 11 12 13 14 15 16 Periods