Assignment 3 ADM 2304 - Applications of Statistical Methods in Business Due Date and Time: 16 March 2025 before 11:59 pm Instructions for Submission 1. You may use StatCrunch, MS Excel Analysis ToolPak for calculations that require technology. However, you must show your manual calculations when asked. 2. For questions that indicate StatCrunch, you must include your StatCrunch output (e.g., screenshots of tables, plots). 3. If you are performing a hypothesis test, make sure you state the hypotheses, the level of significance, the decision criteria/region, the test statistic, and/or the critical value or p-value, if requested), your decision (whether to reject or not to reject the null hypothesis), and a conclusion in managerial terms that answers the question posed. These steps must be completed in addition to any software output. 4. The data for this assignment can be found in the file Assignment 3 Data.xlsx. 5. Your assignment must be typed and uploaded to Brightspace in a single pdf file. The most recent submission prior to the deadline will be graded. Start each question on a different page and answer the questions in order. 6. Late submissions will be accepted according to the late submission policy discussed in class and posted in the course outline on Brightspace. 7. Please do not submit a Declaration of Absence for this assignment. Declaration of Absences apply for exams. 8. Remember to include your integrity statement. Assignments submitted without a signed integrity statement will not be graded. 9. You are not permitted to use AI Large Language Models (LLMs) such as ChatGPT, Claude, Gemini, to complete the assignment. 10. Students who fail to follow any of the instructions above will be penalized with 10% of the marks (for example, if the assignment is marked out of 50, the penalty will be 5 marks). 11. IMPORTANT ADVICE FOR CALCULATIONS: Please keep 3 decimal places for all intermediate steps and final result. 1 Q1: [26] A business analyst would like to study the effect of the age of the car owner on the size of the cash offer for a used car by randomly selecting individuals in three age groups (young, middle, elderly). A 10-year-old car was selected for the study, and cash offers (in hundred dollars) received from different dealers are contained in the data set Assignment 3 data.xlsx. Young Middle Elderly 20 18 32 25 26 35 15 30 35 30 29 40 22 26 28 20 20 29 15 22 10 (a) [2.5] Identify the experimental design, experimental unit/study subject or participant, factor, level, and response (or output) variable. The experimental design seemed to be completely randomized design because the samples of the car owner in each group are random selected. The experimental unit is car owner. The factor is the age group of car owners The levels are young, middle and elderly The response variable is the cash offer for the used car (b) [3] Graph the data using side-by-side box plots. Are the assumptions (i.e., normality and equal variance) of ANOVA met here? Explain. 2 Box plots look symmetric for each group, with no extreme outliers. 3 P>=0.05, the assumption of equal variance is also met (c) [5.5] Perform an appropriate ANOVA test to determine whether there is a difference in average cash offers among 3 different age groups using both critical value and p-value approach. Use α = 0.05 and describe all necessary steps of test of hypothesis (i.e., hypotheses, test statistic, critical value/p-value, decision rule, and decision with justification). Note: You can use directly test statistic and p-value from STATCRUNCH output. (STATCRUNCH) H0: μYoung=μMiddle=μElderly H1: at least one μ is different. 4 F statistic =7.9952 Df(between group_=2 Df (within group)=18 Significance level: α=0.05 Using df(2,18), α=0.05 5 F critical =3.55 Decision rule:(critical value approach) If Fstatistic >Fcritical, reject H0 IF Fstatistic<= Fcritical, fail to reject H0 Since 7.9952>3.55,we reject H0 Decision rule:(p value approach) · From the ANOVA output, p-value = 0.0033. · Decision rule: · If p-value< a(0.05) reject H0 · If p-value>=a(0.05),we fail to reject H0 6 · Since 0.0033 <0.05,we reject H0 · In conclusion we reject H0, which means at a=0.05 at least one μ is different (d) [1.5] Calculate the pooled variance using standard deviations of samples obtained from STATCRUNCH. Compare your pooled variance with the ANOVA output on STATCRUNCH. Does your calculation agree with the STATCRUNCH output? (Manual Calculation and STATCRUNCH) Sp^2=[(n1-1)s1^2+(n2-1)s2^2+(n3-1)s3^2]/(n1+n2+n3-3) =(7*39.69+5*23.37+6*34.4)/18 =33.36 From the Anova table in statcrunch ,the mean square error is also 33.6 which match my manual calculate result. (e) [6] If necessary, determine which groups cash offerings are different from the others and for which groups you cannot confirm, by using Bonferonni adjusted confidence intervals. Assume the family level of significance α = 0.05. (Manual Calculation) aadjusted=0.05/3=0.0167 a/2=0.00835 1-0.00835=0.99165 Comparison 1: Young vs. Middle SE=√33.36*(1/6+1/8)=3.12 ME=(24.833-19.625-2.9618*3.12,24.833-19.625+2.9618*3.12) 7 =(5.208-9.240816,5.208+9.240816) =(-4.03,14.45) Since the interval contains zero, we cannot confirm a significant difference between Young and Middle groups. Comparison 2: Young vs. Elderly: SE:√33.36*(1/7+1/8)=2.99 ME;(31.571-19.625-2.96188*2.99,31.571-19.625+2.96188*2.99) ME(3.09,20.8) Since the interval does not contain zero, there is a significant difference between Young and Elderly groups. Comparison 3: middle vs elderly SE;√33.36*(1/6+1/7)=3.21 ME:(31.571-24.833-2.9618*3.21+31.571-24.833+2.9618*3.21) =(-2.77.16.25) Since the interval contains zero, we cannot confirm a significant difference between Middle and Elderly groups. In summary: Elderly car owners offer significantly higher cash values than young car owners. We cannot confirm significant differences between Young and middle-aged car owners, Middle-aged and elderly car owners. (f) [5] Notwithstanding your answer to part (b), perform Kruskal Wallis nonparametric test at level of significance α = 0.05 to determine whether there is difference among the median cash offers among the 3 age groups using critical value approach. Be sure to sate your hypotheses, test statistic, critical value, decision rule, rank calculation in EXCEL, //and decision with justification. (Manual Calculation and EXCEL) H0: μYoung=μMiddle=μElderly 8 H1: at least one μ is different. α = 0.05 (Critical Value) α = 0.05, df=2, critical value=5.991 decision value: if H> critical value ,reject H0 if H<=critical value, fail to reject H0 H:(12/21*22)*(53^2/8+64^2/6+114^2/7)-3*22 =9.074 because H=9.07>5,991, we reject H0 · conclusion: in the significance level of α = 0.05, there are sufficient evidence to prove there are at least one μ is different 9 (g) [2.5] Perform the Kruskal Wallis nonparametric test using STATCRUNCH, include your output and give your decision using p-value approach. Note: You need to uncheck Adjust for Ties in the STATCRUNCH application. (STATCRUNCH) 10 The output from statcrunch match my manual calculation and excel output. Q2: [24] The Aeroplan program is the frequent flyer program offered by Canada’s largest airline, Air Canada. It allows travelers to earn and redeem points for various travel-related rewards towards flights, hotel, car rental, and merchandise & gifts. The CEO of Air Canada is concerned that many of its loyalty members have accumulated large quantities of free miles and the airline may face difficulties if the redeem requests are submitted around the same time. Hence, a business analyst of Aeroplan program conducted an experiment where four methods of redeeming frequent flyer miles were offered to a random sample of 60 members who had accumulated more than 80,000 frequent flyer miles. The frequent flyer members were equally divided into three age groups. The number of miles redeemed for each frequent flyer member during the 8-week experimentation period are contained in the data set Assignment 3 data.xlsx and also given in the following table. Flights Hotel Car Rental Merchandise and Gifts Under 25 25 to 50 Over 50 30000 0 30000 65000 0 55000 28000 12000 10000 35000 25000 0 40000 70000 0 20000 38000 65000 0 18000 55000 45000 0 40000 0 30000 15000 10000 0 45000 40000 0 35000 60000 25000 38000 38000 48000 30000 50000 50000 40000 45000 25000 20000 0 10000 10000 12000 25000 20000 0 0 30000 0 35000 42000 35000 0 0 Note: You can get the exact critical value from STATCRUNCH/EXCEL or approximate critical value from the table. (a) [4] Identify the experimental design, number of factors along with levels, number of treatments, number of replications, and response (or outcome) variable. 11 Experimental design: two-way ANOVA design Numbers of factors and levels: Factor 1: Age group (3 levels: under 25,25 to 50, over 50) Factor 2: Redemption Method (4 levels: Flights, Hotel, Car Rental, Merchandise and Gifts) Number of Treatments: 3*4=12 treatments Number of Replications: 60/3=20 members per age group 20 ÷ 4 = 5 replications per treatment Response variable: Number of miles redeemed during the 8-week experimentation period (b) [5] Do the data provide sufficient evidence to indicate an interaction between age group and type of redemption? Conduct an appropriate test of hypothesis at 5% level of significance using both critical value and p-value approach.Please provide hypotheses, test statistic, critical value, p-value, decision rule, and decision with justification. (STATCRUNCH) Hypotheses: H₀: There is no interaction between age group and type of redemption H₁: There is an interaction between age group and type of redemption 12 Decision rule: Reject H₀ if F > F-critical. Fail to reject H₀ if F <= F-critical Base on the output ; F=1.2771823 critical(α = 0.05,df₁ = 6,df₂ = 48)= 2.30 Since F = 1.2771823 < 2.30, we fail to reject the null hypothesis. decision rule;Reject H₀ if p-value < α (0.05) ,fail to reject H0 if p value >=a(0.05) Since p-value = 0.2856 > 0.05, we fail to reject the null hypothesis. Using both approaches, we reach the same conclusion: there is insufficient evidence at the 5% significance level to conclude that there is an interaction between age group and type of redemption. (c) [3] Test at 5% level of significance, if the average number of redeemed miles differs among the three age groups. Use the critical value approach. (STATCRUNCH) Hypotheses: H₀: μ₁ = μ₂ = μ₃ H₁: At least one mean is different Significance level: α = 0.05 13 F = 1.7639311 Critical F-value (α = 0.05, df₁ = 2, df₂ = 48) = 3.19 Decision Rule: Reject H₀ if F > F-critical (3.19) Fail to reject H₀ if F ≤ F-critical (3.19) 14 Since F = 1.7639311 < 3.19, we fail to reject the null hypothesis. Using the critical value approach at a 5% significance level, there is not sufficient evidence to conclude that the average number of redeemed miles differs among the three age groups. (d) [3] Test at 5% level of significance, if the average number of redeemed miles differs among the four types of redemption options. Use the critical value approach. (STATCRUNCH) Hypotheses; H₀: μ₁ = μ₂ = μ₃ = μ₄ H₁: At least one mean is different Significance level: α = 0.05 F = 6.1468565(from statcrunch) Critical F-value (α = 0.05, df₁ = 3, df₂ = 48) = 2.80 Decision Rule: Reject H₀ if F > F-critical (2.80) Fail to reject H₀ if F <= F-critical (2.80) Since F = 6.1468565 > 2.80, we reject the null hypothesis. Using the critical value approach at a 5% significance level, there is sufficient evidence to conclude that the average number of redeemed miles differs among the four types of redemption options. (e) [9] Calculate 95% Bonferroni margin of error for the confidence intervals based on all the pairwise differences between the average number of redeemed miles. Now, test the difference between the mean redeemed miles for the following two treatments (Manual Calculation): (i) (Under 25, Hotel) versus (Under 25, Car Rental). g = number of pairwise comparisons = 12C2 = 66 α/2g = 0.05/(2×66) = 0.000379 15 df = 48 MSE = 3.1255e8 n₁ = n₂ = 5 Finding t-critical value for α/(2g) = 0.000379 with df = 48:,t=3.94 Bonferroni =3.94 × √(3.1255e8*2*(1/5)) = 3.94 × √(1.2502e8) =3.94 × 11180.9 =44053 (Under 25, Hotel) vs (Under 25, Car Rental): (41000 -44053,41000+44053) = (-3,053, 85,053) (ii) (Under 25, Hotel) vs (Over 50, Car Rental). =( 44000 - 44053 55600+44053)= (547, 89153) Now, draw two interaction plots (Plot 1: Age (x axis) vs Type (y axis); Plot 2: Type (x axis) vs Age (y axis)) using STATCRUNCH and verify whether you notice the same as you concluded in the above two comparisons (STATCRUNCH). 16 = 17 18 Statement of Academic Integrity Individual Assignment Checklist & Disclosure Please read the disclosure below following the completion of your assignment. Once you have verified these points, hand in this signed disclosure with your assignment. 1. I acknowledge to have read and understood my responsibility for maintaining academic integrity, as defined by the University of Ottawa’s policies and regulations. Furthermore, I understand that any violation of academic integrity may result in strict disciplinary action as outlined in the regulations. 2. I have referenced and/or footnoted all ideas, words, or other intellectual property from other sources used in completing this assignment. 3. A proper bibliography is included, which includes acknowledgement of all sources used to complete this assignment. 4. This is the first time that I have submitted this assignment or essay (either partially or entirely) for academic evaluation. 5. I have not utilized unauthorized assistance or aids including but not limited to outsourcing assignment solutions, and unethical use of online services such as artificial intelligence tools and course-sharing websites. Course Code: Assignment No. / Title: ADM 2304 ASSIGNMENT 1 Use of Plagiarism Yes (Required by Course / Professor) o Detection Tools No (Not Applicable for Type of Assignment) o (e.g., Ouriginal): Date of Submission: Name: Signature: Feb.5th Yilei Wang Yes (Self-Conducted) o No (Not Conducted) o
0
You can add this document to your study collection(s)
Sign in Available only to authorized usersYou can add this document to your saved list
Sign in Available only to authorized users(For complaints, use another form )