competing_on_analytics_notes.txt

Common theme to organizational differentiators of comprehensive analytics competitor:
  Top management has communicated that the organization competes on quantitative, fact-based analysis
  Widespread use of predictive modeling and optimization
  Analyze the heck out of their data
  Comprehensive understanding of their competitors
  Optimize business strategy and planning, can determine impact of unexpected constraint, simulate alternatives
  Model relationship between operational costs and performance
  Can measure lift of intervention strategies and apply results to improve future analysis
  For each subgroup, perform regression analysis to identify factors that most closely correlate with *business event*
  Information-based strategy - Programs operate under common leadership, technology, and tools
Detractors - multiple versions of key indicators within an organization
Enablers - Centralized groups to ensure that critical data and other resources are well managed -> share data without inconsistencies of formats, definitions, and standards

Enterprise Analyst Management - Analysts embedded in different divisions but are centrally managed
Can focus strategically on organization's priorities
Deep resource pool, tell investors, public, and press
CEO Barry Beracha's personal and organizational philosophy "In God we trust. All others bring data."
leadership example: When they need help, they turn to experts who understand the business and how analytics can be applied
  Their analyst lieutenants use plain language and are trusted not to spin the numbers
  
  Analysts: apply technology to multiple business problems, finding the right focus, building the right culture, and hiring the right people to make optimal use of the data they constantly churn
  In end, people and strategy, as much as information technology give such organizations strength.
  Multiattribute utility theory  - a tool for understanding and predicting consumer behaviors and decisions.
  Econometrics - statistical techniques for measuring the lift provided by different ads and promotions over time
  
 Good analysts don't stop at measuring within their networks, they measure their up and down stream as well
 The right culture, measuring, testing, and evaluating quantitative evidence.  
 Rewards culture based on meticulously collected performance measurements
 Senior execs - exhibit hunger for and confidence in fact and analysis
 Like all companies that complete on talent, they pursue the best
 Want them at the table, participating in discussion of the key issues, and determining what information needs the business people have
 Recommend actions to the business partners - be active and critical part of the business unit's success
 predictive modeling, recursive partitioning
 
 Competing on analytics means competing on technology; have a data strategy e.g. snatch data from every conceivable source.

For python, if needed

#install for: asset allocation, portfolio optimization, and risk analytics
#pip install scipy # useful optimizers, SDE
#pip install alphalens # performance and risk analytics
#pip install pyfolio # performance and risk analytics
#pip install Copulalib # risk applications
#pip install sympy # symbol mathematics

install for: pricing instruments

#pip install Quantlib-python

install for: for time series analysis

#pip install pandas # R like features (DataFrame, Series, Panel...)
#pip install statsmodels # from linear regression to sophisticated Dynamic Factor Models
#pip install pandas-datareader # get public source data
#pip install arch # GARCH models

install for: for financial charts

#pip install seaborn
#pip install matplotlib