Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 36 views

    Course Introduction 2018

    Uploaded by

    Jenni Lapiz
    This document provides an overview of the Data Science for Business II course. It discusses moving from predictive to prescriptive modeling and analytics. The course will cover optimization tools for prescriptive modeling, decision analysis, and stochastic programming. Students will work in teams on a project involving either constructing an optimal loan portfolio using loan default predictions, or optimally allocing a betting budget using sports team performance predictions. Team projects will be evaluated based on peer and instructor feedback. Background materials from the prerequisite DSFB1 course are available.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    Course Introduction 2018

    Uploaded by

    Jenni Lapiz
    36 views10 pages
    This document provides an overview of the Data Science for Business II course. It discusses moving from predictive to prescriptive modeling and analytics. The course will cover optimization tools for prescriptive modeling, decision analysis, and stochastic programming. Students will work in teams on a project involving either constructing an optimal loan portfolio using loan default predictions, or optimally allocing a betting budget using sports team performance predictions. Team projects will be evaluated based on peer and instructor feedback. Background materials from the prerequisite DSFB1 course are available.

    Original Description:

    Course Introduction 2018. Credit to Aalto School of business

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides an overview of the Data Science for Business II course. It discusses moving from predictive to prescriptive modeling and analytics. The course will cover optimization tools for prescriptive modeling, decision analysis, and stochastic programming. Students will work in teams on a project involving either constructing an optimal loan portfolio using loan default predictions, or optimally allocing a betting budget using sports team performance predictions. Team projects will be evaluated based on peer and instructor feedback. Background materials from the prerequisite DSFB1 course are available.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    36 views10 pages

    Course Introduction 2018

    Uploaded by

    Jenni Lapiz
    This document provides an overview of the Data Science for Business II course. It discusses moving from predictive to prescriptive modeling and analytics. The course will cover optimization tools for prescriptive modeling, decision analysis, and stochastic programming. Students will work in teams on a project involving either constructing an optimal loan portfolio using loan default predictions, or optimally allocing a betting budget using sports team performance predictions. Team projects will be evaluated based on peer and instructor feedback. Background materials from the prerequisite DSFB1 course are available.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 10

    Data  Science  for  Business  II

    From  predictive  to  prescriptive  modeling

    Pekka Malo, Assist. Prof. (statistics)


    Aalto BIZ / Department of Information and Service Management
    Data  is  dead?

    Data  =  record  of  history  that  


    says  nothing  about  the  
    future  or  “what-­if”  scenarios

    Source:  socialmediatoday.com

    2
    let’s end with a definition
    Companies  need  to  make  decisions

    definition
    alternate definition
    :: analytics is not more data!

    :::: analytics
    analytics isisoptimizing
    finding the optimal
    decision path
    making
    to a desired
    in situations future
    of uncertainty
    Reference:  alfredessa.com

    3
    Path  to  prescription
    3
    Information   What do I want to happen?
    value
    2
    What will happen?
    1 Optimization Strategy

    What happened?

    Risk  forecasting Predictive  modeling

    Data  access Reporting


    Insight

    4
    From  passive  to  active  analytics

    Passive Active
    Descriptive
    Informative Diagnostic
    What  has  
    Why  did  it  happen?
    happened?

    Predictive Prescriptive
    Operative What  will  happen? What  should  I  do?

    5
    Course  timeline  and  topics
    Week  1 Week  2 Week  3 Week  4 Week  5 Week  6

    Case  1:   Visiting  lecture:  


    Recap  on   Introduction  to   Case  2:  
    Prescriptive   From  data  to  
    predictive   prescriptive  

    Course  project  
    presentations
    analytics  in   Sports  analytics business  
    modeling modeling
    Finance opportunities

    • Modern  regression   • Optimization  tools   • Nordic  Investment   • SportsIQ /  Jirka • Jyrki  Koskinen  from  
    analysis  techniques for  prescriptive   Bank  (NIB)  /  Simo   Poropudas Avaamo
    • Regularization  and   modeling Heliövaara • Case  related   Konsultointi
    variable  selection • Decision  analysis   • Case  related   guidance  and   • Time  to  prepare  
    • Forecasting  and   • Fundamentals  of   guidance  and   techniques team  presentations  
    time  series   stochastic   techniques (DL  26.3.)
    programming

    Finance  teams  (final  report  DL  13.4.)  

    Sports  teams  (final  report  DL  20.4.)  

    6
    Grading
    The  course  assessment  is  comprised  of  the  following  two  parts:

    • Class  activity  (tutorials,  lectures,  exercises)   30%


    • Team  case  (course  project) 70%

    All  assignments  must  be  completed  to  pass  the  course.  Evaluation  
    criteria  are  separately  specified  in  each  assignment.

    When  evaluating  work  done  in  teams,  starting  level  of  the  student  
    teams  will  be  taken  into  account  in  grading.  

    Special  attention  is  paid  to  the  teams’  development  in  knowledge  
    sharing  and  learning.

    7
    Team  cases
    Make  sure  you  have  registered  for  one  of  the  two  alternative  
    cases!

    Project  1:  Lending Club  loan  portfolio  construction


    The objective of  the project is  to  (i)  predict the default probabilities of  
    loans in  Lending club  and  (ii)  use these predictions to  construct an  
    optimal loan  portfolio.

    Project  2:  Optimal betting portfolio  for  Liiga  playoffs


    The objective of  the project is  to  (i)  predict the performance of  teams
    in  the playoffs of  the Finnish ice  hockey  league “Liiga”  and  (ii)  use
    these predictions to  optimally allocate a  betting budget.

    8
    Assessment  and  grading  of  team  case  
    (70%  of  total  grade)
    • The  grading  of  team  cases  is  based  on  a  combination  of  peer  
    evaluation  and  a  corresponding  evaluation  by  teachers.  

    • Evaluation  rubrics  will  be  provided  separately.  

    • To  conduct  the  peer  evaluation,  you  will  be  provided  with  a  


    separate  observation  form.  Each  student  will  be  able  to  evaluate  
    each  member  of  the  team.  All  peer  evaluations  will  be  
    confidential.

    9
    Review  material  (DSFB-­1  books)
    If  you  need  a  review  on  the  content  of  DSFB1  course,  which  is  in  pre-­
    requisites,  please  have  a  look  at  the  following  books.

    10

    You might also like