The NESACS Blog
From the Northeastern Section of the ACS, focusing on career management and development
Categories:

Archives:
Meta:
June 2017
S M T W T F S
« May    
 123
45678910
11121314151617
18192021222324
252627282930  
11/27/15
Trends in Technical Careers. Forecasting
Filed under: Mentoring, Mature professionals, Observ. Trends, Alternate Career Paths
Posted by: site admin @ 8:49 am

Been reading Phillip Tetlock’s book “Superforecasters” since predicting
what will happen is something we often like to do in science.  As Charles
Kettering once said about his interest in the future ‘because the rest of
his life’ will be spent there.
Forecasting is hard work and can be learned.  People who do it well,
Tetlock opines, have a strong interest in information, a willingness to
adjust to new data, an ability to synthesize a view from various
perspectives, like a ‘dragonfly’s eye.’  They also pay attention to their
prediction compared to the actual result to learn from.

A rough process outline includes
    1.’unpacking’ the question into components
    2.distinguish between the known and unknown, while not leaving
assumptions untested
    3.assess the question from an objective ‘outside viewpoint’
    4.put the problem into a comparative perspective, which downplays its
uniqueness and treats it as a special case
    5.explore others’ predictions for similarities and differences
    6.pay attention to broader predictions from wider sources [wisdom
of crowds]
   7.synthesize the information and compare to actual, learning what
can be done to improve

Example discussions:
The singularity where technological intelligence overtakes human abilities
is predicted to be in 2030 [Vinge ]  and 2045 [Kurzweil ].
 Interesting competitions in forecasting provide events for evolving
approaches.

3 Responses to “Trends in Technical Careers. Forecasting”

  1. site admin Says:


    Regression to the mean is the flip side to correlation.

    LUCK VS  SKILL in FORECASTING      If you
    are playing chess against a grandmaster, skill dominates. His
    performance is going to correlate very highly with his next
    performance and winning. Thus, there is not going to be much
    regression to the mean. He is going to beat you every time.

    Now, if you are also a grandmaster, two grandmasters facing
    each other, luck will be much more important.
    http://j-dm.org/archives/495

    Illusion of control – totally random process of the toss of
    coin, can use statistics and biases to predict outcomes.

    Illusion of prediction - done by “smart people” where they try
    predictions in other questions can result in a sense that they
    overconfidently predict  outcomes.

     Accountability of predictions—not often followed. Reveals
    much about superforecasters.
       Look at words used for predictions
       Look at time frames of predictions

    Do they:
    Test Alternate hypotheses
    Look for and examine Contrary evidence
    Check for accuracy
    Define what are prediction benchmarks
    Compare predicted outcomes with other forecaster predictions
    Define their calibration
  2. site admin Says:
    http://www.goodreads.com/book/show/23995360-superforecasting
  3. site admin Says:


    Forecasters working in groups can make more accurate predictions
    when certain procedures are used that lessen “groupthink”
    phenomena that small cohesive teams with shared illusions and
    related norms.

    Aggregation of input only works when people form judgments
    independently. Best teams which have been tested need to
       trust each other to be open,
       ask questions about all related topics, theories and evidence,
       not worry about offending others,
                                   the importance and background of other
    team members and
       being willing to work harder and seeing all the data yourself.

Leave a Reply