Lecture 5: Diagnostic Data Analytics

Question 1

Quantitative Models contain:
- ? variables:
> ? variables that can help you make decisions.
> move ? and affect the movement of dependent variables.
- ? variables: included in the model to avoid ?, not relevant to decision making.
- One can’t include all possible variables in her model
=> ? variables: cause errors
- result variable

Answer 1

Quantitative Models contain:
- independent variables:
> decision variables that can help you make decisions.
> move before and affect the movement of dependent variables.
- control variables: included in the model to avoid bias, not relevant to decision making.
- One can’t include all possible variables in her model
=> uncontrollable variables: cause errors
- result variable

Question 2

Algorithm: ?, ? steps

Answer 2

Algorithm: patterns, repeated steps

e.g. I know I’m a morning person because I look at my repeated routines and find out I’m the happiest when I wake up early.

Question 3

Heuristics: finding a good enough solution using ?

e.g. try to drive as fast and as safe as possible while still complying to the rules.

Answer 3

Heuristics: finding a good enough solution using rules

e.g. try to drive as fast and as safe as possible while still complying to the rules.

Question 4

Certainty:

• assume ? knowledge
• all ?? are known
• no surprises!

Answer 4

Certainty:

• assume complete knowledge
• all potential outcomes are known
• no surprises!

Question 5

Uncertainty:

• ? outcomes for each decision
• ? of each outcome is unknown

Answer 5

Uncertainty:

• several outcomes for each decision
• probability of each outcome is unknown

Question 6

Risk analysis (? decision making):

• ? of each outcome is known
• level of ? => Risk (? value)

Answer 6

Risk analysis (Probabilistic decision making):

• probability of each outcome is known
• level of uncertainty => Risk (expected value)

Question 7

Simulation: often used in ?? analysis

Answer 7

Simulation: often used in what-if analysis

Question 8

?: a way of numerically quantifying the relationship between 2 variables.

• variables are highly correlated if the coefficient is > ? (?)

Answer 8

Correlation: a way of numerically quantifying the relationship between 2 variables.

• variables are highly correlated if the coefficient is > 0.8 (80%)