P1.F.4.3 Data Analytics - Analytic Tools

Question 1

Type of Data Analytics

P1.F.4.3 Data Analytics - Analytic Tools

Answer 1

1. Descriptive: Who, what, when and where?
2. Diagnostic: Why something happened?
3. Predictive: What will happen? (forecast)
4. Prescriptive: What should happen? Greatest value because its exercises can lead to decisions that can create value.

Question 2

Predictive Analytic Techniques

P1.F.4.3 Data Analytics - Analytic Tools

Answer 2

1. Find exploratory variables that correlate to dependent variable.
   Example: calculate regression equations
2. Decide which data to include or exclude
   Example: outlier: outside the norm.
3. Derive regression line supported by backtesting
4. Validating the fit: split into two groups; one to derive and one to test
5. Compare with other models

Question 3

Exploratory Data Analysis

P1.F.4.3 Data Analytics - Analytic Tools

Answer 3

1. An exercise undertaken without an existing hypothesis regarding the data.
2. Goal is to find a new and useful relationship among variables

Question 4

Limitations of Data Analytics

P1.F.4.3 Data Analytics - Analytic Tools

Answer 4

1. Doesn’t explain causation or address motives
2. Lacks qualitative measures
3. May encourage transactional focus instead of relationships
4. Doesn’t lead to perfect decisions
5. Confirmation bias must be overcome

Question 5

Data Analytic Model Challenges

P1.F.4.3 Data Analytics - Analytic Tools

Answer 5

1. Will never reconcile exactly
2. Employing the right level of detail
3. Increasing variables increase costs and complexity
4. Randomness always seems present
5. Choosing and sampling population

Question 6

Data Analytic Model Types

P1.F.4.3 Data Analytics - Analytic Tools

Answer 6

1. Clustering: define variables and visually displays them
2. Classification: puts observations into categories
3. Regression: study of relationships among variables
4. Multiple regression: more than one explanatory variable

Question 7

Sensitivity Analysis

P1.F.4.3 Data Analytics - Analytic Tools

Answer 7

1. Refers to the degree to which changes in input variables affect output.
2. Shows which variable are critical and how to measure them.
3. Demonstrates overall quality and data sufficiency
4. Models should be built to accommodate
5. End results demonstrates model trustworthiness

Question 8

Sensitivity Analysis Benefits & Limitations

P1.F.4.3 Data Analytics - Analytic Tools

Answer 8

Benefits

1. Demonstrates model veracity
2. Spotlights important variables to control

Limitations

1. Only shows what to discard
2. Overhead cost that doesn’t add to the value chain

Question 9

Simulation Models

P1.F.4.3 Data Analytics - Analytic Tools

Answer 9

1. Systematic way of dealing with uncertainty
2. Repeatedly test model with randomized inputs
3. Demonstrates range and probability of outputs
4. Vast applications

Question 10

Simulation Model Benefits & Limitations

P1.F.4.3 Data Analytics - Analytic Tools

Answer 10

Benefits

1. Makes decisions in the face of uncertainty.
2. Helpful in replacing intuition, prejudice and flat out guessing
3. Creates confidence around best-case, worst-case and most likely scenarios

Limitations

1. Can’t predict human responses or behaviors to changes
2. Can’t model casual links that affect a particular result in the real world
3. Accuracy depends on input quality

Question 11

What-if & Goal Seeker

P1.F.4.3 Data Analytics - Analytic Tools

Answer 11

1. Both tools to run scenarios to understand possibilities
2. Prepare for best/worst case

What-if

1. Starts with changes in input
2. What will happen if we change this?

Goal-seeking

1. Starts with output goal
2. If we want to change the result, what needs to happen?

Question 12

Regression - Simple & Multiple

P1.F.4.3 Data Analytics - Analytic Tools

Answer 12

1. Find dependent variable
2. From one or more independent (explanatory) variables
3. Contains constants

Simple: one explanatory
Multiple: more than one explanatory

Question 13

Least Squares Line

P1.F.4.3 Data Analytics - Analytic Tools

Answer 13

The line that minimizes the vertical distances between itself and the data points.

Question 14

Least Squares Line Equation

P1.F.4.3 Data Analytics - Analytic Tools

Answer 14

Observed value = Fitted value + Residual

1. Fitted value: vertical line distance between x-axis and the line
2. Observed value: actual point
3. Residual: difference between fitted value and observed value

Question 15

Regression Equation Calculations

P1.F.4.3 Data Analytics - Analytic Tools

Answer 15

y = a + bx

y = the mean y value
a = the optimal y-intercept
b = the optimal slope (variable coefficient)
x = the mean x value

1. b (numerator) = (mean x value - x value) - (mean y value - y value)
2. multiply x difference by y difference
3. b (denominator) = (mean x value - x value) squared
4. add the values

Question 16

Correlation Coefficient: r

P1.F.4.3 Data Analytics - Analytic Tools

Answer 16

1. Correlation between two variables
2. Range from -1 to 1
3. -1 implies perfect inverse correlation. As one variable increases, the other decreases by the same amount.
4. 0 implies no correlation
5. 1 implies perfect correlation

Question 17

Coefficient of Determination: R2

P1.F.4.3 Data Analytics - Analytic Tools

Answer 17

1. Fit between least squares line and observed data
2. % of variance in independent variable explained by least squares line
3. Range from 0% (no explanation) to 100% (perfect explanation)
4. Can’t compare outside context
5. Can be applied to simple and multiple regression

Question 18

Time Series Analysis

P1.F.4.3 Data Analytics - Analytic Tools

Answer 18

1. Values of same variables over time
2. Used in forecasting
3. Trend: generally, are things increasing or decreasing over time, and if so, by how much?
4. Cyclical: how things change over long-term cycles (more than one year)
5. Seasonal: how things change over a one year cycle
6. Irregular: seemingly random and unpredictable, does not repeat in any particular patterns

Question 19

Time Series Analysis Benefits & Limitations

P1.F.4.3 Data Analytics - Analytic Tools

Answer 19

Benefits

1. Assist with understanding decisions
2. Applications almost limitless

Limitations

1. Only shows correlation: don’t help in identifying root cause
2. Echo chamber effect: more useful within range than outside
3. Reliance on lagging indicators (historical observations)
4. Random noise can distort picture