Exam 2 - Chapter 5 - Perform/Types of D.A. Flashcards
AMPS model;
What does the P represent?
Perform the analysis;
Process of evaluating data through analyses to draw conclusions.
What are the four types of accounting data analytics?
- Descriptive Analytics: What happened/is happening?
- Diagnostic Analytics: Why did it happen? Why past results?
- Predictive Analytics: Will it happen in the future?
- Prescriptive analytics; What should we do/How do we optimize our performance?
Describe descriptive analytics
Analytics performed which characterizes, summarizes, and organizes features and properties of data for better understanding.
- Counts
- totals
- sums
- averages
- subtotals
- minimums and maximums
- standard deviations
- ratio analytics
What do these analytics have in common?
They are all examples of descriptive analytics
What are the two main tools to describe descriptive analytics?
- Excel: Pivot Tables and D.A.
- Histograms
Describe diagnostic analytics:
Analytics performed to investigate the underlying cause that cannot be answered by just looking at descriptive data.
What are the two categories of diagnostic analytics?
- Identify anomalies/outliers
- Finding previously unknown linkages, patterns, and relationships
What is the first step of diagnostic analytics?
- What type of analytics is often done at this stage?
Look for and idenitfy unusual, unexpected results/transactions
- Type of D.A.:
Benford’s law
What is Benford’s law?
Explain it.
Type of anlytics to find anomalies/outliers.
In any group of large, randomly produced numbers, there are 9 sets of expected distributions.
- Group 1 has highest density of numbers
- Each successive group has less numbers, down to 9.
- Any numbers outside these 9 groups is an anomaly.
- controls testing
- Excel match() commands
- conditional formatting (to highlight duplicate or unique values)
- graphing and charting (outlier analytics)
- minimums and maximums (to find extreme values)
- fuzzy matching
- data sorts
What do these examples all have in common?
They are all statistical tests for diangostic analytics.
When using diagnostic analytics to find linkages, how might drill down analytics be used?
To uncover additional details by summarizing data at different levels to identify potential correlations between variables.
Example:
Sum accounts receivable by customer and sort highest to lowest to determine uncollectibles.
What are some tests in diagnostic analytics for finding correlations/patterns?
- Drill-down
- Cross-tabulation
- Correlation
- Regression
- Hypothesis testing
How might correlations and regression tests be used in diagnostic analytics?
Correlation: assess relationships between two variables. (adv. exp & sales rev.)
Regression: Assess relationship amongst multiple variables (R&D exp, Capital Exp, and NI)
Desrcibe predicitive analytics:
Analytics performed to provide foresight by i_dentifying patterns in historical data and assessing future probability._
What are the three categories of prescriptive analytics?
- Classification: Separate and classify a sample into two or more groups
- Regression: Predict dependent variable values based on independent variable
- Forecasting using time series analytics: Predict future values based on past
Describe Prescriptive Analytics:
Inquiry performed to identify the best possible options given constraints and changing conditions.
What are the five categories of prescriptive analytics?
- Sensitivity Analytics
- Capital Budgeting
- Marginal analytics
- Goal-seek analytics
- What-if scenario analytics
Prescriptive analytics:
What are sensitivity analytics?
Evaluating outcomes based on multiple possible inputs.
(how sensitive outcomes are to input changes)
Example:
What happens to Bank of America profits if
the interest rates change in the future from
3% to 4% or 5% or more?
Prescriptive Analytics:
What is capital budgeting?
Evaluating future cash flows of capital investments to determine best purchase.
Example
Which assets should we invest in at
Boeing, a new piece of equipment or a
new computer system?
Prescriptive Analytics:
What is marginal analytics?
Determines the change in profit associated with cost/benefit of additional units.
Question example
Should Tesla make or buy its batteries?
Prescriptive Analytics:
What is goal seek analytics?
Form of what if analytics; determines what must be done in order to reach desired result.
Question Example
For McGraw-Hill, what are the needed
sales to break even on sales of this text?
Prescriptive Analytics:
What is what-if scenario analytics?
Analytics of potential future events by considering potential outcomes
Question example
How do we respond to various scenarios
of increased tariffs with China?
Describe the difference in:
- Population
- Sample
- Parameter
- Statistic
Population:
- group of phenomenon having something in common.
- Characteristic data collected on population is called parameter
Sample
- Subset of members of a population selected to represent population
- Data characteristics collected on samples are called statistics
What are probability distributions?
What is a normal distribution?
Function that describes the possible values of random variables and the likelihood that variable will be within a given range.
Normal Distribution
Bell-shaped probability distribution that is symmetric about its mean
What does the z-score tell about a data point?
How many standard deviations (σ) a data point is from its population mean (μ)
What are the differences in:
- Uniform Distribution
- Poisson Distribution
Uniform Distribution
Distribution where all outcomes are equally likely (like a coin toos)
Poisson Distribution
- Mean number of events per interval of time
- Low mean and highly skewed to right