HW4HeatherLangsdon - There could be many causes for this...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
Heather Langsdon Homework Assignment 4 11.105 a. Seven models are fit to the data in step 1; E(y)= β 0 + β 1 x j b. 6 models are fit to the data in step 2; E(y)= β 0 1 x 1 2 x j e. Many t-test have to be conducted and many important terms may have been left out. 13.27 b. 2009: 205.41 Single Exponential Smoothing for CPI Smoothing Constant 0.40 Initial Value 168.032 Sum of Squared Errors (SSE) 4334.63 Mean Squared Error (MSE) 228.138 Standard Error (SE) 15.1042 Mean Absolute Deviation (MAD) 12.8665 Mean Abs Percentage Error (MAPE) 8.09 Mean Percentage Error (MPE) 1.87 Number of Cases 19 Forecast (T) = 205.411 95% C.I. 95% C.I. Lead Lower Bound Forecast Upper Bound 1 175.806 205.411 235.015 13.50 a. The graph of residuals plotted against T shows evidence of positive autocorrelation. This meaning there is evidence of a long positive run in the data.
Background image of page 1
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: There could be many causes for this positive run. One being the more options a customer has from year to year. The market keeps growing giving customers more buyer power. b. At alpha= .05 with P(positive) being .0000, there is sufficient evidence to support the claim that the errors are positively autocorrelated. Student Edition of Statistix 9.0 BUYPOWER, 4/29/2011, 1:55:29 PM Durbin-Watson Test for Autocorrelation Durbin-Watson Statistic 0.0581 P-Values, using Durbin-Watson'S Beta Approximation: P (positive corr) = 0.0000, P (negative corr) = 1.0000 Expected Value of Durbin-Watson Statistic 2.0553 Exact Variance of Durbin-Watson Statistic 0.09930 Cases Included 38 Missing Cases 0...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online