Info3010, Week 8 Notes
Info3010, Week 8 Notes Info3010
Popular in Business Modeling
verified elite notetaker
Popular in Business
verified elite notetaker
This 4 page Class Notes was uploaded by Rebecca Evans on Monday March 7, 2016. The Class Notes belongs to Info3010 at Tulane University taught by Srinivas Krishnamoorthy in Spring 2016. Since its upload, it has received 22 views. For similar materials see Business Modeling in Business at Tulane University.
Reviews for Info3010, Week 8 Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 03/07/16
Monday February 29, 2016: TV Flip Flops A television network earns an average of $1.6M each season from a hit program and loses an average of $0.4M each season on a program that turns out to be a flop. In general, 25% of programs turn out to be hits and 75% turn out to be flops. At a cost of $0.16M, a market research firm will analyze a pilot episode of a prospective program and issue a report predicting whether the given program will end up being a hit. If the program is actually going to be a hit then there is a 90% chance that the market researchers will correctly predict the program to be a hit. If the program is actually going to be a flop then there is an 80% chance that the market researchers will predict it as a flop. Identify the strategy that maximizes this television network’s expected profit in response to a newly proposed television program. Compute the EVSI and EVPI for this decision problem. Notes & Work Base rate of success=25% Conditional probability: probability of an outcome given a condition MR=Market Research P(MR says hit if actual hit)=90% o MR says=outcome o Actual hit=condition P(Actual hit if MR says hit)=construct Bayesian probability table to calculate o Actual hit=outcome o MR says hit=condition Market Research (MR) says Hit Flop 250 Hit (.9)(250) False =225 negative Actual (250-225) =25 Flop False (.8)(75) 750 positive =600 (750-600) =150 375 625 1000 o P(MR says hit)=375/1,000=0.375 o P(Actual hit if MR says hit)=225/375=(hits identified)/(MR says hit)=0.60 DO NOT mix up the 2 probabilities: P(MR says hit if actual hit)=from past data, (Actual hit if MR says hit)=future probability being calculated P(Actual flop if MR says flop)=600/625=(#actual flop if say flop/total # MR say flop)=0.96 Wednesday March 2, 2016: TV Flip Flops Homework Probabilities on decision tree A) EV(a)= (.25*1.6M) + (.75*-0.4M)= $0.1M B) EV(b)=(.04*1.44M) + (.96*-0.56M)= -$0.48M C) (0.6*$1.44M) + (0.4*-$0.56M)=518,400-224,000=$.64M D) EV(d)= (0.375*.64M) + (0.625*- 0.16M)do market research Decision Plan: Do MR If MR says hit then make the show If MR says flop then don’t make the show EVSI EVSI=Expected Value of Sample Information o Sample info is info from a less than perfect test EVSI = EV with free sample info – EV without info EV with free sample information: o EV with free sample information – cost of sample info = EV with paid sample info o EV with free sample information - $0.16M (cost info) = $0.14 (EV w/ paid) EV with free sample info = $0.30M o EV with free sample information= EV with paid information + cost of sample info EVSI=EV with free sample info-EV w/o sample info = $0.30M - $0.1M=$0.2M=EVSI Because the Expected Value of Sample Information is positive number we made the decision to do Market Research Market research=sample information because contains false negatives and false positives (info from less than perfect test) Perfect information= MR Actual Hit flop Hit 250 False negative 0 flop False positive 750 0 Friday March 4, 2016: TV Flip Flops Perfect Information MR says Actual Hit flop Hit 250 False negative 0 flop False positive 750 0 P(MR says Hit)=250/1,000=.25=same as success rate in industry therefore, perfect info Perfect information is unrealistic can know upper bound for how much information is worth EVPI=expected vvalue of perfect info=EV with free perfect info-EV without info=$0.40M - $0.1M=$0.3M It’s the ceiling $ amount we would pay for any information Bayesian analysis A way to calculate revised probabilities based on new information Prior probability (base rate of success) vs. revised probability (get after do market research) o 25% of TV shows are hits (prior) vs. If market research says hit then 60% chance that show will be hit (revised probability) New info can be o Expert opinion o Marketing study o Scientific test Tip: think in terms of frequencies to find revised probability Fundamental question: o What is the value of info that improves our decision making? Types of info: o Sample info: from less than perfect test o Perfect info: info from perfect test, never actually available, provides upper bound on value of any info Expected value of sample information o EVSI=EV with free sample info - EV without info Expected value of perfect information o EVPI=EV w/ free perfect info -EV without info
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'