berkshire student file

B C D E F G H M Berkshire Computer Sales Berkshire Computer Sales is a small company located in Oldenburg, Washington The founder of the company, Jonathan Berkshire, began the business by selling computer systems through mail-order at discount prices. Berkshire was one of the first computer mail-order companies to offer a tollfree phone number to their customers for support and trouble shooting. While the company has grown over time, many new competitors have entered the computer mail-order marketplace so that Berkshire's share of this market has actually declined Five years ago, Bob Gravenstein a marketing consultant was contracted by Berkshire to develop long-term marketing plans and strategies for the company. After careful study, he recommended that Berkshire branch out to reach a new and growing segment of the computer sales market college students. At that time, most universities were providing microcomputer labs and facilities for their students. However, many students were purchasing their own computers and printers so that they could have access to a computer at any time in their apartments or dorms. After graduation, students take the computers, with which they are already familiar, to their new jobs or use them at home after work hours. The percentage of college students owning a computer was increasing rapidly and Bob Gravenstein recommended that Berkshire Computer Sales take advantage of this marketing opportunity The marketing plan developed by Bob Gravenstein worked as follows: Five universities were initially selected for the program with expansion to other universities planned over time. Any student in at least his or her sophomore year at one of these universities was eligible to purchase a discounted JCN 2001 microcomputer system with printer from Berkshire Computer Sales under the program. The ICN-2001 is a private label, fully compatible system with all of the features of brand name models. The student makes a small payment each semester at the same time regular tuition payments were due. When the student graduates and finds a job, the payments are increased so that the computer will be paid for within two years after graduation. If the student fails to make payments at any time, Berkshire could repossess the computer system. The prospect of future sales of computer equipment to these students was also part of the marketing strategy. The JCN-2001 is an entry level computer system that suffices for most academic and professional uses. Eventually, however, many students who purchased this system would outgrow it and require an upgrade to a more powerful machine with more features. Bob Gravenstein argued that after their good experience with the JCN-2001, these customers would make their future purchases from Berkshire Computer Company Today, five years later, Berkshire Computer Sales is still operating the student purchase program and has expanded it to several more universities. There is currently enough data available from the early days of the program for Berkshire to determine whether or not the program has been successful and if it should be continued. To discuss the future of the student purchase program, Jonathan Berkshire has called a meeting with Ben Davis, who has been in charge of the program since it began, and Teresa Grimes, the new vice-president of marketing Berkshire called you both in here today to discuss the student purchase program. As you know the program has been in place for approximately five years and we need to make decisions about where to go from here. Ben, weren't you telling me last week that we now have enough data to evaluate the program? Davis "Yes sir. Any student who began the program as a sophomore five years ago should be out of school and employed for at least two years. Berkshire "Well, based on your information would you say the program has been a success or a failure?" Davis "That's a tough call, sir. While most of the participants in the program eventually pay their account in full, we have had a high rate of defaults, some while the students were still in school, but mostly after they graduated 0 fx B D E G H J L currently enough data available from the early days of the program for Berkshire to determine whether or not the program has been successful and it should be continued. To discuss the future of the student purchase program, Jonathan Berkshire has called a meeting with Ben Davis, who has been in chance of the program since it began, and Teresa Grimes, the new vice-president of marketing Berkshire "I called you both in here today to discuss the student purchase program. As you know, the program has been in place for approximately five years and we need to make decisions about where to go from here. Ben, weren't you telling me last week that we now have enough data to evaluate the program? Davis "Yes, r. Any student who began the program as a sophomore five years ago should be out of school and employed for at least two years. "Well , based on your information, would you say the program has been a success or a fallure? Berahi Davis *That's a tough call, sir. While most of the participants in the program eventually pay their account in full, we have had a high rate of defaults, some while the students were still in school, but mostly after they graduated." Berkshire "How much are we losing on those who default?" Davis Each case is different. As I said, some default early and others after they have graduated. There are also the costs of repossession and repair to bring the product back up to resale quality in many instances we were not able to retrieve the computer systems. Our data suggest that our average loss on each student customer who defaults is about $1,200. On the other hand, our average profit from participants who pay their accounts in full is approximately 5750. Overall, we are close to just breaking even." Grimes "hen, have you considered qualifying students for the program, much like a loan officer would quality someone for a loan Davis: "We had initially thought about doing that, but we didn't believe that there was much information, if any, in the way of a credit history on most college students. Applicants are still requested to provide as much information as possible on their application, including their class grade point average, work experience, scholarshipsand how much of their college expenses were earned through work. However, we weren't sure that this information was particularly useful for screening applicants." Grimer *Knowing that we were going to have this discussion today. I had one of my assistants wo is well-versed in statistics look aver some of those data last week. She has come up with a screening test based only on the information you have been collecting from student applicants. By being more selective about whom we allow to partidpate in the student purchase program, it may be possible to increase our profit from the program Davis it would be easy enough to check out her screening test by trying it out on our early data from the program. In those cases, we know whether or not the student actually defaulted." Berkshire Why don't the two of you spend some time on this idea and get back to me next week? At that time I want a recommendation to either discontinue the program continue it as is or continue it using this screening test'idea. Make sure you have the evidence to back up your recommendation Case Questions 1. Estimate the probability that a customer participating in the student purchase program will default. Compute the expected profit from selling the JCN-2001 system to a university student under the existing program. Has the student purchase program been profitable? Explain and attach supporting output. 2. Based on the data, is defaulting on the computer payments independent of the university attended by the student? Show your work. (Hint: Realize that this is only a sample of data.) If there were dependence, what might be possible causes for this dependence? 3. Assume that 70 is a passing score on the screening test. Estimate the probability that a student who passes the screening test will eventually default Compute the expected profit from selling the JCN-2001 system to a university student who passes the screening test. Is the screening test used in this fashion significantly better than not using the test at all? Explain your answer and attach any supporting output. 4. Can you find a better way of using the screening test that results in higher expected profits for Berkshire Computer Sales? Explain your solution and attach any relevant supporting output. 5. Make your final recommendation for the student Purchase program to Jonathan Berkshire. Include any modifications that you would make to the program and discuss the potential changes caused by these modifications. Also discuss any other issues that should be considered. Data Description The file contains data on all participants in the student purchase program who by now should have either paid in full or defaulted (i.e., those participants who should have graduated and held a job for at least two years). A partial listing of the data is shown below. These data are coded as follows: Student: Student transaction number for identification purposes. School: University where the student was enrolled. Default: 1, in the event of default: 0, if account was paid in full on time. When: 1, if default occurred (or paid in full) before graduation; o, if default occurred (or paid in full) after graduation Score on screening test based on student applicant information such as his or her class, grade point average, work experience, scholarships, and how much of their college expenses were earned through work. Score: 19-an 3.2.2- E Arlal 10 ' ' 1 Wrap art General Pula E 23 Merge & Co % ) Subscription Required to edit and Save Start your free one month trial of Microsoft 365 or into activate an enting subscription H21 Cand Formal Cal Formatting estable styles Delle Format D When M Default 1 ID 2 3 4 What type of variable is? 1 School 5547 4503 1219 1843 6607 6386 4427 0668 3897 6456 2 2 4 1 4 2 1 2 2 School Default Score Score as pass or fail a Tomta formating the VP 70 FloScore>60 passar) 1 D 1 4350 8 0 10 11 12 19 14 15 16 17 18 10 0 D 0 7833 4600 BADS 1005 8127 5400 3301 0878 4 2 4 3 3 2 5 5 2 0 D Score Pass 70 TAI 0 58 fal D 52 fail 0 56 foil O 47 Toll 0 sa fal D 50 fall 03 Fail 71 pass 1 61 l 0 72 pass 34 fail 1 38 Fail o 35 pass 0 61 0 80 fail 41 a 0 67 fail 69 54 tail 0 52 fail 36 foil 1 61 0 50 fall 0 58 al 0 50 tal 0 53 fa 02 fall 0 52 fail 1 06 fail 0 65 0 48 0 pas 77 pass o fal 0 53 fal 35 F mil Cautions Data Description 2327 23 24 25 1 1 0 1 1 1 0 2 1 3 aaaa 7165 22:10 4542 2517 7600 5623 4054 5027 4854 1270 2474 6571 7 9540 6614 389 2644 ARA Ascenen D 0 1 0 0 1 3 1 5 5 0 3 Ideen PA Delicate Batman Summary Store Cas Background Ready scription Required to Edit and Save Start your free one month trial of Microsoft 365 or Sign in to activate anisting subscription Formatting as Table Styles Form A EST Screening Store fel ZONE 4 1.00 ww 63 TE et for 2 + 1 . ta 000 1 4 3100 1 9 . tar 2000 TOP + + 1 tul HERE How would you describe the distribution of store! + WIE Series Store Wales SF 0 2 1 WE # 1 Summary for some Canon Data e motion Baiteroute Canced Al fx D B. Dalad | les PEN PP 4 2 al Whether school? ENCE School Schools independer New Deeper PA RE PANAN HAPA Wind TERECETRERTISIERESSEERT ETTER DEERE CERT INDOVI School School 2 Schools School 42 17 104 52 14 542 137 of one Dersom Numero Probability of School 1 given default PrSchool 1 Default) Count Column Label Row Labels School School No De School School Set 35 be 52 62 205 22 Grand Total 12 67 85 104 32 34 342 Total Total Column Label RowLahden School 1 32 School 2 33 School 42 School 17 Schools Grand Total 137 23 24% 30.66% 12.41% 2.42% 100005 32 230 3124.00% 42 10.50 17 1241 13249 t tous 17 18 School Default P[School 1 Default) 32/137.0 23 12 10 27 28 20 30 31 32 33 34 35 36 37 ill 2,315 No Default Detail Default Only 20 40 41 42 45 46 47 Because "Default" has already occurred, we are only looking at the subpopulation of students who have detaulted. The denominator for P(School XlDefault) is 137. 40 50 Probability of Default ghen School 1 - PDefault School 1 Denomi Numrator Default School 53 54 59 Cound School Coun School No Dell Default OrTotal Grand Total 35 838 981 Default No Default Dufauit Grand Total 15 52 65 62 42 104 35 17 52 21 12 14 205 137 342 SONE School School Schools Grand Total 60 02 63 Petauit even School 1) PDefault School 1) 10 No Default Dett Default Only Because "Default" has already occurred, we are only looking at the subpopulation of students who have defaulted. The denominator for P(School XlDefault) is 137. Probability of Default oliven School 1.Defrukt school 1) Den G School Daterials Na full Data Default Ord Total Beho Ne Duted Det Ganda 12 School Schools ch Schools Grand 35 42 17 62 35 23 305 85 104 52 34 37 Default given School 1)- Petal School 1) P(DefaultSchool) PINO DetaultSchool 1) 32/674 /67-52% New w 32, 48% 35 35,52% Sched Scho School 1 only Because "School 1" has already occurred, we are only looking at the subpopulation of students who are in School 1. The denominatorfor PDefault School 1) is 67. B C D E F G H M Berkshire Computer Sales Berkshire Computer Sales is a small company located in Oldenburg, Washington The founder of the company, Jonathan Berkshire, began the business by selling computer systems through mail-order at discount prices. Berkshire was one of the first computer mail-order companies to offer a tollfree phone number to their customers for support and trouble shooting. While the company has grown over time, many new competitors have entered the computer mail-order marketplace so that Berkshire's share of this market has actually declined Five years ago, Bob Gravenstein a marketing consultant was contracted by Berkshire to develop long-term marketing plans and strategies for the company. After careful study, he recommended that Berkshire branch out to reach a new and growing segment of the computer sales market college students. At that time, most universities were providing microcomputer labs and facilities for their students. However, many students were purchasing their own computers and printers so that they could have access to a computer at any time in their apartments or dorms. After graduation, students take the computers, with which they are already familiar, to their new jobs or use them at home after work hours. The percentage of college students owning a computer was increasing rapidly and Bob Gravenstein recommended that Berkshire Computer Sales take advantage of this marketing opportunity The marketing plan developed by Bob Gravenstein worked as follows: Five universities were initially selected for the program with expansion to other universities planned over time. Any student in at least his or her sophomore year at one of these universities was eligible to purchase a discounted JCN 2001 microcomputer system with printer from Berkshire Computer Sales under the program. The ICN-2001 is a private label, fully compatible system with all of the features of brand name models. The student makes a small payment each semester at the same time regular tuition payments were due. When the student graduates and finds a job, the payments are increased so that the computer will be paid for within two years after graduation. If the student fails to make payments at any time, Berkshire could repossess the computer system. The prospect of future sales of computer equipment to these students was also part of the marketing strategy. The JCN-2001 is an entry level computer system that suffices for most academic and professional uses. Eventually, however, many students who purchased this system would outgrow it and require an upgrade to a more powerful machine with more features. Bob Gravenstein argued that after their good experience with the JCN-2001, these customers would make their future purchases from Berkshire Computer Company Today, five years later, Berkshire Computer Sales is still operating the student purchase program and has expanded it to several more universities. There is currently enough data available from the early days of the program for Berkshire to determine whether or not the program has been successful and if it should be continued. To discuss the future of the student purchase program, Jonathan Berkshire has called a meeting with Ben Davis, who has been in charge of the program since it began, and Teresa Grimes, the new vice-president of marketing Berkshire called you both in here today to discuss the student purchase program. As you know the program has been in place for approximately five years and we need to make decisions about where to go from here. Ben, weren't you telling me last week that we now have enough data to evaluate the program? Davis "Yes sir. Any student who began the program as a sophomore five years ago should be out of school and employed for at least two years. Berkshire "Well, based on your information would you say the program has been a success or a failure?" Davis "That's a tough call, sir. While most of the participants in the program eventually pay their account in full, we have had a high rate of defaults, some while the students were still in school, but mostly after they graduated 0 fx B D E G H J L currently enough data available from the early days of the program for Berkshire to determine whether or not the program has been successful and it should be continued. To discuss the future of the student purchase program, Jonathan Berkshire has called a meeting with Ben Davis, who has been in chance of the program since it began, and Teresa Grimes, the new vice-president of marketing Berkshire "I called you both in here today to discuss the student purchase program. As you know, the program has been in place for approximately five years and we need to make decisions about where to go from here. Ben, weren't you telling me last week that we now have enough data to evaluate the program? Davis "Yes, r. Any student who began the program as a sophomore five years ago should be out of school and employed for at least two years. "Well , based on your information, would you say the program has been a success or a fallure? Berahi Davis *That's a tough call, sir. While most of the participants in the program eventually pay their account in full, we have had a high rate of defaults, some while the students were still in school, but mostly after they graduated." Berkshire "How much are we losing on those who default?" Davis Each case is different. As I said, some default early and others after they have graduated. There are also the costs of repossession and repair to bring the product back up to resale quality in many instances we were not able to retrieve the computer systems. Our data suggest that our average loss on each student customer who defaults is about $1,200. On the other hand, our average profit from participants who pay their accounts in full is approximately 5750. Overall, we are close to just breaking even." Grimes "hen, have you considered qualifying students for the program, much like a loan officer would quality someone for a loan Davis: "We had initially thought about doing that, but we didn't believe that there was much information, if any, in the way of a credit history on most college students. Applicants are still requested to provide as much information as possible on their application, including their class grade point average, work experience, scholarshipsand how much of their college expenses were earned through work. However, we weren't sure that this information was particularly useful for screening applicants." Grimer *Knowing that we were going to have this discussion today. I had one of my assistants wo is well-versed in statistics look aver some of those data last week. She has come up with a screening test based only on the information you have been collecting from student applicants. By being more selective about whom we allow to partidpate in the student purchase program, it may be possible to increase our profit from the program Davis it would be easy enough to check out her screening test by trying it out on our early data from the program. In those cases, we know whether or not the student actually defaulted." Berkshire Why don't the two of you spend some time on this idea and get back to me next week? At that time I want a recommendation to either discontinue the program continue it as is or continue it using this screening test'idea. Make sure you have the evidence to back up your recommendation Case Questions 1. Estimate the probability that a customer participating in the student purchase program will default. Compute the expected profit from selling the JCN-2001 system to a university student under the existing program. Has the student purchase program been profitable? Explain and attach supporting output. 2. Based on the data, is defaulting on the computer payments independent of the university attended by the student? Show your work. (Hint: Realize that this is only a sample of data.) If there were dependence, what might be possible causes for this dependence? 3. Assume that 70 is a passing score on the screening test. Estimate the probability that a student who passes the screening test will eventually default Compute the expected profit from selling the JCN-2001 system to a university student who passes the screening test. Is the screening test used in this fashion significantly better than not using the test at all? Explain your answer and attach any supporting output. 4. Can you find a better way of using the screening test that results in higher expected profits for Berkshire Computer Sales? Explain your solution and attach any relevant supporting output. 5. Make your final recommendation for the student Purchase program to Jonathan Berkshire. Include any modifications that you would make to the program and discuss the potential changes caused by these modifications. Also discuss any other issues that should be considered. Data Description The file contains data on all participants in the student purchase program who by now should have either paid in full or defaulted (i.e., those participants who should have graduated and held a job for at least two years). A partial listing of the data is shown below. These data are coded as follows: Student: Student transaction number for identification purposes. School: University where the student was enrolled. Default: 1, in the event of default: 0, if account was paid in full on time. When: 1, if default occurred (or paid in full) before graduation; o, if default occurred (or paid in full) after graduation Score on screening test based on student applicant information such as his or her class, grade point average, work experience, scholarships, and how much of their college expenses were earned through work. Score: 19-an 3.2.2- E Arlal 10 ' ' 1 Wrap art General Pula E 23 Merge & Co % ) Subscription Required to edit and Save Start your free one month trial of Microsoft 365 or into activate an enting subscription H21 Cand Formal Cal Formatting estable styles Delle Format D When M Default 1 ID 2 3 4 What type of variable is? 1 School 5547 4503 1219 1843 6607 6386 4427 0668 3897 6456 2 2 4 1 4 2 1 2 2 School Default Score Score as pass or fail a Tomta formating the VP 70 FloScore>60 passar) 1 D 1 4350 8 0 10 11 12 19 14 15 16 17 18 10 0 D 0 7833 4600 BADS 1005 8127 5400 3301 0878 4 2 4 3 3 2 5 5 2 0 D Score Pass 70 TAI 0 58 fal D 52 fail 0 56 foil O 47 Toll 0 sa fal D 50 fall 03 Fail 71 pass 1 61 l 0 72 pass 34 fail 1 38 Fail o 35 pass 0 61 0 80 fail 41 a 0 67 fail 69 54 tail 0 52 fail 36 foil 1 61 0 50 fall 0 58 al 0 50 tal 0 53 fa 02 fall 0 52 fail 1 06 fail 0 65 0 48 0 pas 77 pass o fal 0 53 fal 35 F mil Cautions Data Description 2327 23 24 25 1 1 0 1 1 1 0 2 1 3 aaaa 7165 22:10 4542 2517 7600 5623 4054 5027 4854 1270 2474 6571 7 9540 6614 389 2644 ARA Ascenen D 0 1 0 0 1 3 1 5 5 0 3 Ideen PA Delicate Batman Summary Store Cas Background Ready scription Required to Edit and Save Start your free one month trial of Microsoft 365 or Sign in to activate anisting subscription Formatting as Table Styles Form A EST Screening Store fel ZONE 4 1.00 ww 63 TE et for 2 + 1 . ta 000 1 4 3100 1 9 . tar 2000 TOP + + 1 tul HERE How would you describe the distribution of store! + WIE Series Store Wales SF 0 2 1 WE # 1 Summary for some Canon Data e motion Baiteroute Canced Al fx D B. Dalad | les PEN PP 4 2 al Whether school? ENCE School Schools independer New Deeper PA RE PANAN HAPA Wind TERECETRERTISIERESSEERT ETTER DEERE CERT INDOVI School School 2 Schools School 42 17 104 52 14 542 137 of one Dersom Numero Probability of School 1 given default PrSchool 1 Default) Count Column Label Row Labels School School No De School School Set 35 be 52 62 205 22 Grand Total 12 67 85 104 32 34 342 Total Total Column Label RowLahden School 1 32 School 2 33 School 42 School 17 Schools Grand Total 137 23 24% 30.66% 12.41% 2.42% 100005 32 230 3124.00% 42 10.50 17 1241 13249 t tous 17 18 School Default P[School 1 Default) 32/137.0 23 12 10 27 28 20 30 31 32 33 34 35 36 37 ill 2,315 No Default Detail Default Only 20 40 41 42 45 46 47 Because "Default" has already occurred, we are only looking at the subpopulation of students who have detaulted. The denominator for P(School XlDefault) is 137. 40 50 Probability of Default ghen School 1 - PDefault School 1 Denomi Numrator Default School 53 54 59 Cound School Coun School No Dell Default OrTotal Grand Total 35 838 981 Default No Default Dufauit Grand Total 15 52 65 62 42 104 35 17 52 21 12 14 205 137 342 SONE School School Schools Grand Total 60 02 63 Petauit even School 1) PDefault School 1) 10 No Default Dett Default Only Because "Default" has already occurred, we are only looking at the subpopulation of students who have defaulted. The denominator for P(School XlDefault) is 137. Probability of Default oliven School 1.Defrukt school 1) Den G School Daterials Na full Data Default Ord Total Beho Ne Duted Det Ganda 12 School Schools ch Schools Grand 35 42 17 62 35 23 305 85 104 52 34 37 Default given School 1)- Petal School 1) P(DefaultSchool) PINO DetaultSchool 1) 32/674 /67-52% New w 32, 48% 35 35,52% Sched Scho School 1 only Because "School 1" has already occurred, we are only looking at the subpopulation of students who are in School 1. The denominatorfor PDefault School 1) is 67