i need the expertise to do the assignment in a very precious way please and give the justification got his answers.
below is thee guidelines :1. Group Assignment (30%): Following the class, you will be assigned a newspaper article which uses data analytics to make a claim and an implicit recommendation. Your assignment will be to critique this article based on the 5-point check-list we learned in class: you will go through each item on your checklist and explain how the “symptom” does or doesn’t apply to the claims/recommendation made in the article.
Deliverable: Max 2-page bullet-point submission, addressing each point on your checklist, explain why the answer to the items on your checklist is “yes” or “no”.
1/21/2020 An MBA is still a great boost for salaries | Financial Times https://www.ft.com/content/a2813230-d69b-11e7-a303-9060cb1e5f44 1/3 FT Series MBA ranking preview 2018 MBA fees are increasing but they still represent an opportunity for a substantial salary bump for those who complete their studies © FT montage; Dreamstime Laurent Ortmans JANUARY 3 2018 A top MBA is a significant investment. The programme for Stanford Graduate School of Business costs more than $145,000 in tuition fees. Students tell us that the main reason they study for an MBA qualification is to increase their salary. Personal development, interpersonal skills and ethics matter to them, but students still expect a return on investment. MBA students had a tougher ride after the financial crisis of 2008. Between 2008 and 2014, the average salary of MBA graduates three years after they left business school increased by just 4 per cent to $127,000. At the same time, the average cost of two-year MBAs increased by 44 per cent to $104,000. MBA An MBA is still a great boost for salaries Average pay for alumni is $142,000 - up 12 per cent since 2014 https://www.ft.com/mba-ranking-preview https://www.ft.com/stream/c6ffcc92-b711-30a2-bdc9-258dc7e377ae http://rankings.ft.com/businessschoolrankings/stanford-university-gsb https://www.ft.com/mba 1/21/2020 An MBA is still a great boost for salaries | Financial Times https://www.ft.com/content/a2813230-d69b-11e7-a303-9060cb1e5f44 2/3 252% The record for the highest salary increase, set by Chicago Booth’s MBA alumni Salaries have since picked up to reach an average of $142,000 in 2017, up 12 per cent on 2014, and early data analysis from the 2018 ranking indicate that salaries are still increasing. Average alumni salaries have increased year on year for about three-quarters of schools. Average salaries increased by $7,000 in 2017, the largest increase in absolute terms in more than 12 years. All sectors apart from three — education, transport and logistics and law — saw average salaries increase. The biggest year-on-year increases were in the healthcare and industrial sectors, both up 10 per cent to average about $143,000. Salaries in these two sectors are down in 2018. If the latest trend continues, the overall average salary should be near to $150,000 in 2018. We also measure salary increases over pre-MBA levels. Every year up to 2014 and since we began collecting data, MBA graduates always at least doubled their salaries within three years of completing their degree. Alumni from Chicago’s Booth School of Business hold the record for the highest salary jump since our records began: up 252 per cent in the 2002 ranking during the heyday of the dotcom boom. But the overall average salary boost fell for the first time below 100 per cent to 93 per cent in 2015 as average pre-MBA salaries increased much faster than post-MBA salaries. That same year, http://rankings.ft.com/businessschoolrankings/university-of-chicago-gsb 1/21/2020 An MBA is still a great boost for salaries | Financial Times https://www.ft.com/content/a2813230-d69b-11e7-a303-9060cb1e5f44 3/3 Copyright The Financial Times Limited 2020. All rights reserved. alumni from Stanford Graduate School of Business had their lowest salary increase at only 80 per cent. The global average salary increase moved back above 100 per cent again in 2017. Just over half of all ranked schools had salary increases greater than 100 per cent. In 2018, almost two-thirds of alumni cohorts more than doubled their salaries. The FT ranking of global MBA courses will be published on January 29 http://help.ft.com/help/legal-privacy/copyright/copyright-policy/ Part2_good_bad_analytics Part 2: How to tell Good Analytics from Bad Analytics Data Analytics for Managers Raji Jayaraman Professor of Economics, ESMT Berlin Working knowledge of data analytics 1. Tell good analytics from bad analytics 2. Understand what descriptive, predictive and prescriptive analytics are 3. Use data analytics to understand problems 4. Evaluate whether or not a solution “works” Pentathlon Case 1. Put yourself in the shoes of Anna. What is her major concern? Make a convincing case for promotional email limit, using the evidence you have generated. 2. Put yourself in the shoes of Francois. What is his major concern? Make a convincing case against promotional email limit, using the evidence you have generated. 3. Put yourself in the shoes of the CEO. Given the evidence presented to you by Anna and Francois, would you impose a limit on promotional email activity or not? Anna’s case for a promotional email limit Francois’ case against a promotional email limit “Online ads are effective. Online manufacturing ads are effective. Together, they are (almost) doubly effective.” Source: The 2016 Deloitte Millennial Survey Source: Forbes Insights Survey of Business Executives YOU need to decide: Is this good analytics or bad analytics? • Am I interpreting the data correctly? • Have I come to the right conclusion given the evidence at hand? • Am I making the right decision? …How do you figure this out? Diagnosing bad analytics always starts with a simple question: “What is the data-generating process?” “How did I end up seeing what I am seeing in the data analysis?” 5-point checklist Symptom # 1: Bad data Is this bad data? − Is it representative of the population I am interested in studying? − Has the data been measured and collected accurately? Symptom # 2: Omitted variables Are there omitted variables? − Is there something else that I haven’t accounted for (z), that is correlated with my “explanation” (x) that can explain the outcomes (y) I observe in the analytics? − There is some factor z, where z à x and z à y Symptom # 3: Selection bias Is there selection bias? - Are the units (e.g. people) systematically different in one group than in another? Symptom # 4: Reverse causality Is there reverse causality? − I am acting as if the explanatory variables have a causal effect on the outcomes, but does causality actually run the other way, from “outcome” to “explanation”? − It is not x à y. It is y à x Symptom # 5: Mutual dependence or simultaneity Is there mutual dependence or simultaneity? − Is the “explanation” causing the “outcome” and the “outcome” causing the “explanation”? • x à y and y à x − Are x and y jointly (i.e. simultaneously) determined? Procedure for diagnosing bad analytics Ask: “What is the data-generating process?” 1. Bad data? 2. Omitted variables? 3. Selection bias? 4. Reverse causality? 5. Mutual dependence or simultaneity? NOTE: These are not mutually exclusive! You can answer “yes” to more than one question. If you have NONE of these symptoms: you are in good to go in terms of recommending an action. If you have ANY one of these symptoms, stop: you need to be careful about recommending an action (the obvious interpretation may not be the correct one.) ? If the answer is “not sure”: talk to a domain expert Golden Rule # 2: Diagnose bad analytics by asking: “What is the data-generating process?” 1. Bad data? 2. Omitted variables? 3. Selection bias? 4. Reverse causality? 5. Mutual dependence or simultaneity? “Firefighter Death Squad” 19 0 10000 20000 30000 40000 50000 60000 0 10 20 30 40 50 60 Fi re D am ag e (€ ) Number ofFirefighters “Reverse Causality” June August June July May July August “Omitted Variables” “Ice Cream Kills” 20 “Tutoring Pays Off” 21 85 60 0 10 20 30 40 50 60 70 80 90 Tutor NoTutor TestScore(%) “Selection Bias” “R&D drives revenues” 22 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11 20 12 20 13 20 14 20 15 20 16 20 17 Revenues and R&D expenditures R&D Expenditure (€ 100K's) Revenue (€ Mil.) “Mutual Dependence” “Simultaneity” “Online ads are effective. Online manufacturing ads are effective. Together, they are (almost) doubly effective.” What is the data generating process? Who ends up seeing these ads and who doesn’t? 1. Is this bad data? 2. Are there omitted variables? 3. Is there selection bias? 4. Is there reverse causality? 5. Is there mutual dependence/simultaneity? Source: The 2016 Deloitte Millennial Survey “We collected the views of nearly 7,700 Millennials representing 29 countries around the globe. All participants were born after 1982, have obtained a college or university degree, are employed fulltime, and predominantly work in large (100+ employees), private sector organizations." What is the data generating process? Who stays in a company and who leaves? Who gets support (or perceives the organization as being supportive), who doesn’t? 1. Is this bad data? 2. Are there omitted variables? 3. Is there selection bias? 4. Is there reverse causality? 5. Is there mutual dependence/simultaneity? “Diversity is a Driver of Innovation” Source: Forbes Insights (2011): ”Fostering Innovation through a Diverse Workforce”. https://images.forbes.com/forbesinsights/StudyPDFs/Innovation_Through_Diversity.pdf “The information in this report is based on the results of a survey and one-on-one interviews conducted by Forbes Insights. Forbes Insights surveyed 321 executives with direct responsibility or oversight for their companies’ diversity and inclusion programs. All respondents worked for large global enterprises with annual revenues of more than US$500 million.“ What is the data generating process? Who is responding to this question? 1. Is this bad data? 2. Are there omitted variables? 3. Is there selection bias? 4. Is there reverse causality? 5. Is there mutual dependence/simultaneity? https://images.forbes.com/forbesinsights/StudyPDFs/Innovation_Through_Diversity.pdf Remember the Pentathlon Case? Can you reconcile Elena and Hugo’s evidence? Elena’s evidence for email limits Hugo’s evidence against email limits “What is the data-generating process?” “Who gets promotional emails and who doesn’t?” Golden Rule # 2: Diagnose bad analytics by asking: “What is the data-generating process?” 1. Bad data? 2. Omitted variables? 3. Selection bias? 4. Reverse causality? 5. Mutual dependence or simultaneity?