Ch 4 Demand Estimation

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Chapter 4Demand Estimation Marketing Research ApproachesRegression AnalysisSimple & Multiple Reg AnalysisDemand Estimation by Reg Analysis

pp. 137-183

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Demand Estimation:Marketing Research ApproachesConsumer Surveys: questioning people w/ questionnaires + trained interviewers p.141Observational Research: gathering information by watching them buying products p. 141Consumer Clinics: laboratory experiments; participants are given a sum of money & asked to spend it reaction on changes in price, packaging, displays, etc p. 142Market Experiments: conducted in actual market place or several markets record the responses consumers p. 144Virtual Shopping: a virtual store stimulated on the computer screen consumers touch its image (3D modeling) p. 146Virtual Management: more sophisticated model (2002) using computational models + data base, econometrics, information technology pp. 73 & 146

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Regression AnalysisRegression Line: Line of Best Fit

Regression Line: Minimizes the sum of the squared vertical deviations (et) of each point from the regression line.

Ordinary Least Squares (OLS) Method

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Ordinary Least Squares (OLS)Model: et = Vertical deviation/ Residual/error

= expected valuep.149

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Ordinary Least Squares (OLS)Objective: Determine the slope and intercept that minimize the sum of the squared errors.

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Tests of SignificanceStandard Error of the Slope Estimate(n k) = degree of freedomn = 10; k = 2 (= parameters and b)df = 10 2 = 8

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Tests of SignificanceCalculation of the t StatisticDegrees of Freedom = (n-k) = (10-2) = 8Critical Value at 5% level =2.306 6.79 > 2.306 a significant relationship between X and Y2 parameterst = calculated valuetabular value

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Tests of SignificanceDecomposition of Sum of SquaresTotal Variation = Explained Variation + Unexplained Variation

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Tests of SignificanceCoefficient of DeterminationR = 0 to 1R = 0 (none of variation of Y were explained by the variation in X)p. 15785% of the total variation in the firms sales is accounted for the variation in the firms Advertising expenditures

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Tests of SignificanceCoefficient of Correlationr = 92% means that variables X and Y vary together 92% of the time.If r = -1 all the sample observation points fall on a negatively sloped straightline the sign of r is always the same as the sign of b^ (the estimated slope of coefficient)

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Multiple Regression AnalysisAnalysis of Variance and F StatisticCalculated value F = 46.61 (F statistic) is greater than critical value in the table of F distribution (Appendix C) = 4.74 significant relationshippp. 161-164; Table 4.7= 46.61

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Problems in Regression AnalysisMulticollinearity: Two or more explanatory variables are highly correlated insignificant even though R very highHeteroskedasticity: Variance of error term is not independent of the Y variable: e as X e should be constant p. 166Autocorrelation: Consecutive error/ residual terms are correlated: time series data- missing variable Durbin Watson test p. 166-167See: pp 165-168

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Durbin-Watson StatisticTest for AutocorrelationIf d = 2, autocorrelation is absent.p.167

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Steps in Demand EstimationModel Specification: Identify VariablesCollect DataSpecify Functional FormEstimate FunctionTest the Results

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *Functional Form SpecificationsLinear Function:Power Function:Estimation Format:

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide * Hubungan tabungan (S), neraca perdagangan (TB), pinjaman luar negeri (F) terhadap pertumbuhan (g):

g = 2,929 + 0,02996 S - 0,0312 TB - 0,287 F (0,939) (-1,453) (-4,959)

F = 21,745 R = 0,916Contoh 1:

Prepared by Robert F. Brooker, Ph.D. Copyright 2004 by South-Western, a division of Thomson Learning. All rights reserved.Slide *HASIL ESTIMASIit = 18,926 + 0,39 t 1,476 yt 0,14et + 0.099et-1 (2,092) (-3,586) (-1,546) (1,024)

R2 = 0,552F = 6,784 DW = 0.423963 it = tingkat bunga jangka pendek yang digunakan oleh bank sentralt = laju inflasi tahun tyt = pertumbuhan PDB tahun tet = nilai tukar riil pada tahun tet-1 = nilai tukar riil pada tahun t-1f dan g = koefisien bentuk dasar dari Taylor type ruleh0 dan h1 = koefisien dari nilai tukar riil yang menjadi pengembangan dari Taylor type rule

Persamaan dari Taylor type ruleContoh 2:

Managerial Economics: Chapter 4Managerial Economics: Chapter 4January 2012Improved by Nurzaman Bachtiar*Improved by Nurzaman BachtiarManagerial Economics: Chapter 4Managerial Economics: Chapter 4January 2012Improved by Nurzaman Bachtiar*Improved by Nurzaman Bachtiar

Ch 4 Demand Estimation

Documents