2006 STAT3613 Past Paper 1

pdf

School

The University of Hong Kong *

*We aren’t endorsed by this school

Course

3613

Subject

Statistics

Date

Nov 24, 2024

Type

pdf

Pages

6

Uploaded by KidNeutron12719

Report
THE UNIVERSITY OF HONG KONG DEPARTMENT OF STATISTICS AND ACTUARIAL SCIENCE STAT2313 MARKETING ENGINEERING December 23, 2006 Time: 9:30a.m. -11:30 a.m. Candidates taking examinations that permit the use of calculators may use any calculator which fulfils the following criteria: (a) it should be self-contained, silent, battery-operated and pocket-sized and (b) it should have numeral-display facilities only and should be used only for the purposes of calculation. It is the candidate's responsibility to ensure that the calculator operates satisfactorily and the candidate must record the name and type of the calculator on the front page of the examination scripts. Lists of permitted/prohibited calculators will not be made available to candidates for reference, and the onus will be on the candidate to ensure that the calculator used will not be in violation of the criteria listed above. Answer ALL FOUR questions. The questions are of equal value. 1. The information of a promotion resource allocation exercise that Company A performed is given in the table. The price of the product is $3.7 and the cost is $3.2. Promotional spending ($million) 0 0.52 1 1.56 2.21 2.73 Sales (million units) 5.6 6.75 8 10.83 12.4 13.23 The ADBUDG model is employed to study the data given as xc Y=a+b--- Xc+d where Y is the market response, X is the marketing activity, and a, b, c, and dare the parameters. Suggested initial estimates for the parameters are a = lower bound of the response curve b =(upper bound)- (lower bound of the response curve) Suppose X= 1 and Y= 1 d=-b--1 1-a Y'(d+IY c = where Y' = the 1st derivative of Ywith respect to X (slope) of the bxd response curve at X= 1 1
STA T2313 Marketing Engineering (a) (b) (c) Normalize the sales with reference level at promotional spending= $1 million. Calculate the initial estimates based on the provided formulae. Substitute the initial estimates in (b) into the ADBUDG model, (i) Find the predicted values for the normalized sales for the various amounts of (ii) (iii) (iv) promotional spending given in the table. Find the error sum of squares. FindR 2 Calculate the net profit for the various amounts of promotional spending given in the table. (v) Based on (iv), plot the net profit against the various amounts of promotional spending. (vi) Estimate from the plot the optimal promotional spending so as to maximize the net profit. 2. Six products (A, B, C, D, E, and F) from different companies are ranked by a number of customers in terms of 5 attributes - Product quality, Overall service, Delivery speed, Salsesforce image, and Price flexibility. The overall ranking of the products are obtained as a contingency table. Correspondence analysis is employed to study the data. Parts of the SPS S output are provided as follows. Summary Confidence Singular Prooortion of Inertia Value Singular Standard Correlation Dimension Value Inertia Chi Square Sig. Accounted for Cumulative Deviation 2 1 .422 .178 .714 .714 .051 .007 2 .255 .065 .261 .975 .056 3 .066 .004 .018 .992 4 .043 .002 .008 1.000 Total .249 72.191 .oooa 1.000 1.000 a. 20 degrees of freedom Overview Row PointS' Score in Dimension Contribution Of Point to Inertia of Dimension Of Dimension to Inertia of Point Row Mass 1 2 Inertia 1 2 1 2 Total Product quality .138 1.380 -.280 .114 .623 .043 .970 .024 .995 Overall service .200 -.055 .598 .020 .001 .281 .013 .911 .924 Delivery speed .245 .251 .174 .011 .037 .029 .573 .167 .740 Salsesforce image .228 -.664 .210 .046 .238 .039 .924 .056 .979 Price flexibility .190 -.473 -.904 .058 .100 .608 .310 .686 .996 Active Total 1.000 .249 1.000 1.000 a. Symmetrical normalization 2
·. STA T2313 Marketing Engineering Overview Column PointS' Score in Dimension Contribution Of Point to Inertia of Dimension Of Dimension to Inertia of Point Column Mass 1 2 Inertia 1 2 1 2 Total A .186 -.358 .593 .028 .057 .257 .365 .603 .968 B .145 -.689 -.488 .038 .163 .135 .766 .232 .998 c .152 .733 -.641 .051 .194 .244 .668 .308 .976 D .183 -.245 .501 .018 .026 .180 .259 .653 .911 E .193 .971 .207 .080 .432 .032 .961 .026 .987 F .141 -.619 -.523 .034 .129 .152 .669 .288 .958 Active Total 1.000 .249 1.000 1.000 a. Symmetrical normalization Confidence Column Points Confidence Row Points Standard Deviation in Standard Deviation in Dimension Correlation Dimension Correlation Column 1 2 1-2 Row 1 2 1-2 A .218 .206 .-269 Product quality .160 .291 .454 B .161 .142 -.763 Overall service .242 .201 .044 c .277 .219 .464 Delivery speed .192 .215 -.140 D .208 .228 .193 Salsesforce image .137 .223 .305 E .155 .237 -.453 Price flexibility .280 .158 -.494 F .201 .231 -.154 (a) What is proportion of the total variation explained by each of the 4 dimensions and what is the proportion of the total variation explained by all 4 dimensions? (b) A researcher suggests that the 2-dimensional solution is adequate. Justify the suggestion. (c) Based on the 2-dimensional solution, (i) Comment on the fitness of the model. (ii) Interpret the 2 dimensions in terms of the 5 attributes. (iii) Interpret the 2 dimensions in terms of the 6 products. (iv) Construct the perceptual map. (v) How many segments are suggested by the perceptual map? (vi) Describe the segments. 3
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help
ST A T23I3 Marketing Engineering 3. A study is conducted for the types of emotional responses to consumption experience based on a sample of owners of newly purchased cars. I25 questionnaires were completed regarding respondents' emotional reactions to their automobiles. The questionnaire 1 consists of IO variables:- X1: Interest; X2: Joy; X3: Surprise; X4: Sadness; X5: Anger; X6: Disgust; X7: Contempt; X8: Fear; X9: Shame; XI 0: Guilt. A Principal component factor analysis was applied to the standardized values. The eigenvalues are given as Eigenvalue I 5.87I3 2 1.42I7 3 0.9538 4 0.4984 5 0.43I6 6 0.2642 7 O.I863 8 O.I800 9 O.I376 10 0.0550 (a) Determine the number of factors based on (i) the scree plot. (ii) latent root criterion. (b) A 2-factor model was employed and the initialloadings are given as Factor I Factor 2 XI 0.3I3I 0.4765 X2 -0.1950 0.8847 X3 0.5073 0.5358 X4 0.8577 -0.2249 X5 0.8720 -O.I429 X6 0.9395 -O.II77 X7 0.9460 -0.0438 X8 0.8862 O.I329 X9 0.8II4 0.1427 - XIO 0.8723 -0.0144 (i) Interpret the factors. (ii) Calculate and comment on the communalities. (c) V arimax rotation was applied and the rotated loadings for the first rotated factor is given as Factor I XI 0.2II7 X2 -0.3676 X3 0.390I X4 0.8853 X5 0.8830 X6 0.9441 X7 0.9357 X8 0.84I8 X9 0.7666 XlO 0.8576 4
ST AT2313 Marketing Engineering (i) Find the loadings of the second rotated factor. (ii) Interpret the rotated loadings. (iii) Calculate the variance explained by each rotated factor. (d) The background information of the respondents are also collected which are the age (in years), educational level (as primary or below, secondary, or above secondary) and income (in dollars). Suggest statistical method(s) to study the relationship between the 2 factors and the background information. 4. A researcher studies the switching behaviour of a number of consumers across 5 different brands and types of instant coffee. The table below shows the transaction matrix that . reports the switching behaviour for the consumers. To Brand A Brand B Brand B Brand C Brand D decaffeinated caffeinated decaffeinated caffeinated caffeinated regular freeze dried freeze dried regular regular Brand A decaffeinated 93 7 17 19 18 regular Brand B caffeinated 9 80 12 11 24 freeze dried Brand B From decaffeinated 9 14 46 3 7 freeze dried Brand C caffeinated 19 18 4 82 29 regular Brand D caffeinated 26 11 6 35 184 regular For example, that the number of times we observe a switch from Brand A decaffeinated regular to Brand B caffeinated freeze dried is 7 and the number of times we observe a switch from Brand A decaffeinated regular to Brand B decaffeinated freeze dried is 17. The researcher claims that the transaction matrix could be used to reflect the level of competition between the instant coffee. Let Pif be the proportion of the consumers who selected product i but switched to productj. The proposed measure of competition between product i and productj is given as (pu + pji) /2. (a) Justify the researcher's claim. (b) Calculate the values ofpij for all products. (c) Calculate the proximity matrix based on the researcher's proposed formula. Is it a similarity or dissimilarity matrix? (d) A single-linkage hierarchical cluster analysis by the build-up method was employed to study the data based on the proposed proximity matrix in (c). The icicle graph is provided. 5
STAT2313 Marketing Engineering Vertical Icicle Case CD N CD C) CD N C) C) CD CD CD CD CD .... .::: ~ .... .... 11::: 11::: 11::: 11::: C1S C1S 11::: C1S C1S 0 0 C1S 0 0 CD CD 0 Cl u "C "C cc aO ~ ~ cc N ..... Number of clusters ~ 1 X X X X X X X X X 2 X X X X X X X X 3 X X X X X X X 4 X X X X X X (i) Report the cluster solution at each stage of the build-up method. (ii) Find the value of (dis)similarity of concern to join two clusters at each stage. (iii) Sketch a dendrogram for the cluster analysis. The distances are not required to be in scale. (iv) Describe the competitive structure of the instant coffee market. **************END OF PAPER************** 6
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help