c For each data point i12n find z i using Equation 103 then find the average of

C for each data point i12n find z i using equation

This preview shows page 51 - 54 out of 66 pages.

c. For each data point i=1,2,..,n, find z i using Equation (10.3), then find the average of z. d. Use the average z in ŷ = LV + z * DV L=2 . This is the Nonlinear Regression function. Example 10.1: Nonlinear Multiple Variable Regression Using Z-Theory Consider Example 9.2 with two independent variables, x 1 and x 2 , and one dependent variable, y. Information for the past seven periods is also given in Table 10.1. Compare the quality of the solution obtained by Z-theory versus the solution obtained by the conventional Quadratic Regression method. The quadratic solution obtained in Example 9.2 was, b 0 = 33.003, b 1,1 = -0.554, b 1,2 = 0.082, b 2,1 = -1.467, and b 2,2 = 0.028. Therefore, LV= 33.003 + (– 0.554 x 1 + 0.082 x 1 2 ) + (– 1.467 x 2 + 0.028 x 2 2 ) where v 1 (x 1 ) = (– 0.554 x 1 + 0.082 x 1 2 ) , and v 2 (x 2 ) = (– 1.467 x 2 + 0.028 x 2 2 ) DV L=2 = The following table shows LV and DV for each data point. For example, for point i=1, LV 1 =33.003 + (– 0.554*9 + 0.082*9 2 ) + (– 1.467*31 + 0.028*31 2 ) = 16.09 DV 1 == 37.54. z 1 = (16.03-15.72)/37.54= -0.008. The average z= 0.004. Therefore, the regression function is: N.B. The information contained within is the sole intellectual property of Professor Behnam Malakooti of Case Western Reserve University, Cleveland, Ohio. All rights reserved. No part may be reproduced without prior written permission. Page 51 of 66
Image of page 51
Professor Behnam Malakooti Forecasting Revision 49-5 June 10, 2012 ŷ = LV + 0.004* DV L=2 i x 1,i x 2,i y i Quadratic Regression- (Prob.9.2) , ŷ i = LV v 1 (x 1 ) v 2 (x 2 ) DV z ŷ Error 1 9 31 15.72 16.03 1.66 - 18.57 37.47 - 0.008 16.19 0.47 2 5 42 20.1 6 20.16 -0.72 - 12.22 38.53 0.000 20.32 0.16 3 1 35 15.3 15.3 -0.47 - 17.05 35.99 0.000 15.45 0.15 4 6 31 14.34 14.01 -0.37 - 18.57 35.61 0.009 14.16 0.18 5 4 48 24.8 4 24.84 -0.90 -5.90 40.10 0.000 25.01 0.17 6 7 51 31.57 31.03 0.14 -1.99 45.22 0.012 31.22 0.35 7 2 45 23.4 7 22.82 -0.78 -9.32 39.87 0.016 22.99 0.48 - Avg z= 0.004 MAD = 0.28 Table 10.1 : Example 10.1 data and Z-nonlinear regression solution Now, we can compare Nonlinear Z-utility function to Quadratic function solution by comparing their MAD. To compare the two methods, find the MAD. The following results show that Z- utility function provided a better solution. Function MAD 52
Image of page 52
Professor Behnam Malakooti Forecasting Revision 49-5 June 10, 2012 Quadratic Regression LV=33.003 – 0.554 x 1 + 0.082 x 1 2 – 1.467 x 2 + 0.028 x 2 2 0.48 Z-utility Nonlinear Regression ŷ =LV +0.004DV 0.28 10.2 Nonlinear Time Series by Z-Theory* The approach for solving time series forecasting is based on Section 10.1 Nonlinear Regression. For this problem, only two variables are considered. The first independent variable is, t, time, where x t = t for t=1, 2,…, n. The second independent variable is predicted F t for t = 1, 2,…, n, by using any of given Time-Series methods (e.g. using Exponential Smoothing (Section 4.1) or Linear Regression (Section 7.2) or Quadratic Regression Section 9.1). These two independent variables are evaluated by the actual output, y t , e.g. the actual demand y t =D t . First, we find F t for t = 1, 2,…, n using a forecasting method. Then, we find the Quadratic Regression (10.8) for the given two independent variables, x t and F t , i.e. LV = b 0 +( b 11 *t + b 12 *t 2 )+( b 21 *F t + b 22 *F t 2 ) = b 0 +v 1 (t) +v 2 (F t ) (10.8) Example 10.2: Nonlinear Regression Time Series Using Linear Regression Consider Example 7.2 as shown in Table 7.2. Solving this problem by Linear Regression method of Section 7.2 (alternatively, one can use Quadratic Regression method of Section 9.1), the solution is: F t = 1320 + 510*t. Table 10.2 shows the details.
Image of page 53
Image of page 54

You've reached the end of your free preview.

Want to read all 66 pages?

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture