SOA EXAM PA - FORMULA SHEET
QUESTIONS AND ANSWERS
What are the three main categories of predictive modeling problems? - Correct Answers
-Descriptive, Predictive, Prescriptive
What are the focus and aim of Descriptive Modeling Problems? - Correct Answers -
Focus: What happened in the Past
Aim: To describe or interpret observed trends by identifying relationships between
variables
What are the focus and aim of Predictive Modeling Problems? - Correct Answers -
Focus: What will happen in the future
Aim: To make accurate predictions
What are the focus and aim of Prescriptive Modeling Problems? - Correct Answers -
Focus: The impacts of different prescribed decisions
Aim: To answer the what if and what is the best course of action questions
What are characteristics of predictive modeling problems? - Correct Answers -issue,
questions, data, impact better solution, update
What is the "issue" characteristic of predictive modeling problems? - Correct Answers -
There is a clearly identified and defined business issue to be addressed
What is the "questions" characteristic of predictive modeling problems? - Correct
Answers -The issue can be addressed with a few well defined questions
What is the "data" characteristic of predictive modeling problems? - Correct Answers -
Good and useful data is available for answering the questions outlined in previous
characteristic
What is the "impact" characteristic of predictive modeling problems? - Correct Answers -
The predictions will likely drive actions or increase understanding
What is the "better solution" characteristic of predictive modeling problems? - Correct
Answers -Predictive analytics likely produces a solution better than the existing
approach
, What is the "update" characteristic of predictive modeling problems? - Correct Answers
-We can continue to monitor and update the models when new data becomes available
What is a General Strategy to produce a meaningful problem definition? - Correct
Answers -Get to the root cause of the business issue and make it specific enough to be
solvable
What are Specific Strategies to produce a meaningful problem definition? - Correct
Answers -Hypothesis and Key Performance Indicators (KPIs)
What is the "hypothesis" specific strategy to produce a meaningful problem definition? -
Correct Answers -Use prior knowledge of the business problem to ask questions and
develop testable hypotheses
What is the "KPIs" specific strategy to produce a meaningful problem definition? -
Correct Answers -Select appropriate key performance indicators to provide a
quantitative basis for measuring success
What are some constraints of finding the problem definition? - Correct Answers -The
availability of easily accessible and high quality data. Implementation issue, eg the
presence of necessary IT infrastructure and technology to fit complex models efficiently,
the cost and effort required to maintain the selected model
Key features of Data Design - Correct Answers -Relevance, sampling, granularity
What is the definition of relevance in reference to data design? - Correct Answers -Need
to ensure that the data is unbiased, ie., representative of the environment where the
model will operate.
Population - Important for the data source to be a good proxy of the true population of
interest
Time Frame - Choose a time period which best reflects the business environment of
interest. In general, recent history is better than distant history
What is the definition of Sampling in reference to data design? - Correct Answers -The
process of taking a subset of observations from the data source to generate the dataset
Random Sampling - Randomly draw observations from the underlying population
without replacement. Each record is equally likely to be sampled.
Stratified Sampling - Divide the underlying population onto a number of non-overlapping
strata (often w.r.t target) non-randomly, then randomly sample a set number of
observations from each stratum to get a more representative sample.
QUESTIONS AND ANSWERS
What are the three main categories of predictive modeling problems? - Correct Answers
-Descriptive, Predictive, Prescriptive
What are the focus and aim of Descriptive Modeling Problems? - Correct Answers -
Focus: What happened in the Past
Aim: To describe or interpret observed trends by identifying relationships between
variables
What are the focus and aim of Predictive Modeling Problems? - Correct Answers -
Focus: What will happen in the future
Aim: To make accurate predictions
What are the focus and aim of Prescriptive Modeling Problems? - Correct Answers -
Focus: The impacts of different prescribed decisions
Aim: To answer the what if and what is the best course of action questions
What are characteristics of predictive modeling problems? - Correct Answers -issue,
questions, data, impact better solution, update
What is the "issue" characteristic of predictive modeling problems? - Correct Answers -
There is a clearly identified and defined business issue to be addressed
What is the "questions" characteristic of predictive modeling problems? - Correct
Answers -The issue can be addressed with a few well defined questions
What is the "data" characteristic of predictive modeling problems? - Correct Answers -
Good and useful data is available for answering the questions outlined in previous
characteristic
What is the "impact" characteristic of predictive modeling problems? - Correct Answers -
The predictions will likely drive actions or increase understanding
What is the "better solution" characteristic of predictive modeling problems? - Correct
Answers -Predictive analytics likely produces a solution better than the existing
approach
, What is the "update" characteristic of predictive modeling problems? - Correct Answers
-We can continue to monitor and update the models when new data becomes available
What is a General Strategy to produce a meaningful problem definition? - Correct
Answers -Get to the root cause of the business issue and make it specific enough to be
solvable
What are Specific Strategies to produce a meaningful problem definition? - Correct
Answers -Hypothesis and Key Performance Indicators (KPIs)
What is the "hypothesis" specific strategy to produce a meaningful problem definition? -
Correct Answers -Use prior knowledge of the business problem to ask questions and
develop testable hypotheses
What is the "KPIs" specific strategy to produce a meaningful problem definition? -
Correct Answers -Select appropriate key performance indicators to provide a
quantitative basis for measuring success
What are some constraints of finding the problem definition? - Correct Answers -The
availability of easily accessible and high quality data. Implementation issue, eg the
presence of necessary IT infrastructure and technology to fit complex models efficiently,
the cost and effort required to maintain the selected model
Key features of Data Design - Correct Answers -Relevance, sampling, granularity
What is the definition of relevance in reference to data design? - Correct Answers -Need
to ensure that the data is unbiased, ie., representative of the environment where the
model will operate.
Population - Important for the data source to be a good proxy of the true population of
interest
Time Frame - Choose a time period which best reflects the business environment of
interest. In general, recent history is better than distant history
What is the definition of Sampling in reference to data design? - Correct Answers -The
process of taking a subset of observations from the data source to generate the dataset
Random Sampling - Randomly draw observations from the underlying population
without replacement. Each record is equally likely to be sampled.
Stratified Sampling - Divide the underlying population onto a number of non-overlapping
strata (often w.r.t target) non-randomly, then randomly sample a set number of
observations from each stratum to get a more representative sample.
