Midterm Flashcards
Ch 1-4
1
Q
Analytics uses..
A
- Data
- Information Technology
- Statistical Analysis
- Quantitative Methods
- Mathematical or Computer based models
2
Q
Analytics Purpose
A
to help managers gain improved insight about their business operations and make better, fact- based decisions
3
Q
Analytics Applications
A
- Pricing
- Customer Segmentation
- Merchandising
- Location
- Social Media
4
Q
BA has strong relationship with…
A
- profitability
- revenue
- shareholder return
5
Q
3 Kinds of Analytics
A
- Descriptive Analytics
- Predictive Analytics
- Prescriptive Analytics
6
Q
Descriptive Analytics
A
- Uses data to understand past and present
- Metrics and measures
7
Q
Predictive Analytics
A
- Analyzes past performance in an effort to predict the future
- Forecasting, Simulation
8
Q
Prescriptive Analytics
A
- Uses optimization techniques to identify the best alternative to maximize or minimize some objective
- operations research or management science.
9
Q
Data
A
collected fact and figures
10
Q
Database
A
- collection of computer files containing data
- data are interlinked by fields or attributes that are common across files
11
Q
Information
A
comes from analyzing, organizing, and transforming data
12
Q
Big Data
A
Massive amounts of business data from a wide variety of sources
- most available in real time
- can be uncertain or unpredictable
13
Q
Big Data datasets generated by
A
web applications, social networks, click streams, sensors, and cards
14
Q
Big Data helps
A
organizations better understand and predict customer behavior and improve customer service
15
Q
Big Data Characteristics
A
- Volume
- Variety
- Velocity
- Veracity
16
Q
Volume
A
- amount of data increases
- Big today, Bigger tomorrow
17
Q
Variety
A
- Many sources
- Unstructured and messy
18
Q
Velocity
A
Captured in real time and quickly incorporated into decisions
19
Q
Veracity
A
- How reliable is the data?
- Need high-quality data and understanding
20
Q
Metrics
A
are used to quantify performance
21
Q
Measures
A
are numerical value of metrics
22
Q
2 types of Descriptive Analytics Metrics
A
- Discrete Metrics
2. Continuous Metrics
23
Q
Discrete Metrics
A
- Involves counting
- Ex) Number or proportion of on time deliveries, incorrect or incomplete orders, # of errors in an invoice
24
Q
Continuous Metrics
A
- measured on a contunuum
- ex) Delivery time, package weight, purchase price
25
Metrics data can be one of 4 types
1. Categorical (nominal) data
2. Ordinal data
3. Interval data
4. Ratio data
26
Categorical (Nominal) Data
categorized according to a specified characteristic that bear no quantitative relationship
ex) location, employee job
27
Ordinal Data
Data that is ranked or ordered according to some relationship with one another, no fixed units of measurements
ex) college rankings, survey responses
28
Interval Data
- Ordinal data but with constant differences between observations
- No true zero point
- Ratios are not meaningful
Ex)Temperature readings, SAT scores, Time
29
Ratio Data
- Continuous values and have a natural zero point
- Ratios are meaningful
- Decimals
Ex) Monthly sales
30
Models
An abstraction or representation of a real system, idea, or object
-Captures the most important features
31
Forms of Models
1. Written or verbal description
2. Visual display
3. Mathematical formula
4. Spreadsheet representation
32
Decision Models
A decision model is a model used to understand, analyze, or facilitate decision making
33
Types of Model Input & Output
```
Inputs:
1. Data
2. Uncontrollable variables
3. Decision variables (controllable)
Outputs:
1. Performance measures
2. Behavioral measures
```
34
Types of Model Input & Output
```
Inputs:
1. Data
2. Uncontrollable variables
3. Decision variables (controllable)
Outputs:
1. Performance measures
2. Behavioral measures
```
35
Influence Diagrams
Influence diagrams visually show how various model elements relate to each other
(circles, squares, directed arc)
36
Circles
Variables that cannot be controlled
37
Squares
Variables that can be controlled, or decisions
38
Directed arc
One node influences the other
39
Predictive Decision Models
- often incorporate uncertainty to help managers analyze risk
- Aim to predict what will happen in the future
- Use data from the past to define a potential future relationship (cause/ effect)
- Usually a mathematical model but can also be in written, visual, or spreadsheet forms
40
Prescriptive Decision Models
Prescriptive decision models help decision makers identify the best solution
- optimization
- Objective Function
- Constraints
- Optimal Solution
41
Optimization
Finding values of decision variables that minimize (or maximize) something such as cost (or profit)
42
Objective function
The equation that minimizes (or maximizes) the quantity of interest
43
Constraints
Limitations or restrictions
44
Optimal solution
Values of the decision variables at the minimum (or maximum) point
45
Optimal solution
Values of the decision variables at the minimum (or maximum) point
46
Six steps in the problem solving process
1. Recognizing the problem
2. Defining the problem
3. Structuring the problem
4. Analyzing the problem
5. Interpreting results and making a decision
6. Implementing the solution
47
Recognizing the Problem
- Realize the problem exists!
- Problems exist when there is a gap between what is happening and what we think should be happening
Ex) Costs are too high compared with competitors
48
Defining the Problem
- Clearly defining the problem is not a trivial task!
- Must separate the problem from the symptom
- Complexity increases when the following occur
(Large number of courses of action, Several competing objectives, External groups are affected, Problem owner and problem solver are not the same person, Time constraints exist, Problem belongs to a group vs. an individual
49
Structuring the Problem
- Stating goals and objectives
- Characterizing the possible decisions
- Identifying any constraints or restrictions
- Develop a formal model
50
Analyzing the Problem
- Identifying and applying appropriate BA techniques
- Typically involves experimentation, statistical analysis, or a solution process
- Evaluate scenarios, analyze risks associated with alternatives, meet goals, or final optimal solution
Focus for this class!
51
Interpreting Results and Making a Decision
- Managers interpret the results from the analysis phase
- Incorporate subjective judgment as needed
- Understand limitations and model assumptions
- Make a decision utilizing the above information
52
Interpreting Results and Making a Decision
- Managers interpret the results from the analysis phase
- Incorporate subjective judgment as needed
- Understand limitations and model assumptions
- Make a decision utilizing the above information
53
Implementing the Solution
- Translate the results of the model back to the real world
- Make the solution work in the organization by providing adequate training and resources
- Manage change, build trust, etc.
- Inform and balance needs of the stakeholders (Customers, suppliers employees)
- Implementation could be a project!
54
Data Visualization
is the process of displaying data (often in large quantities) in a meaningful fashion to provide insights to support better decisions
(tables vs graphs)
55
Data Visualization allows us to
analyze current and past data, which may reveal patterns and relationships
56
Tables and graphs
improve communication and understanding so Senior management is able to quickly understand the message and impact
57
Dashboards
is a visual representation of a set of key business measures
| - Term related to an automobile’s control panel
58
Dashboards provide
important summaries of key business information
59
Dashboards help
manage a business process or function
60
Graphs and Charts
- Take raw data and display it in a picture
- Map out data associated with variables to determine patterns and relationships
- Communication tool for both model results and data exploration
- One of the most common descriptive analytics
61
6 Types of Charts
1. Column/bar
2. Line
3. Pie
4. Area
5. Scatter
6. Bubble
62
Column/ Bar Charts
show current and past performance to compare categorical data
63
Column/ Bar Chart Forms
1. Clustered: compares values across categories
2. Stacked: displays contribution of each variable towards the overall total
3. 100% Stacked: compares contribution of each variable towards the total as a percent
64
Line Charts
show trends in data over time
| Ex) us exports to china
65
Pie Charts
- display the relative proportion of each data source to the total
- Works well for a few variables
Ex) Marital status
66
Area Charts
- combine pie and line charts
- Shows trends as well as the percentage of the total for each variable
- few variables
Ex) energy consumptions
67
Scatter Charts
- charts show the relationship between two variables
- Need data in pairs to construct
Ex) house size vs market value
68
Bubble Charts
- are a form of scatter charts
- Size of data marker corresponds to the value of a third variable
- Plotting three variables in two dimensions
Ex) Price, price/earnings ratio, market cap
69
Pareto Analysis
- involves sorting data and calculating cumulative proportions
- Isolates variables and determines the main drivers
- Provides focus for continuous improvement initiatives
70
Statistic
is a summary measure of data
71
Descriptive statistics
are methods that describe and summarize data
72
Frequency Distributions
Often displayed as a table that shows the number of observations in each group
- groups are non-overlapping
- “Mutually exclusive and collectively exhaustive”
- usually categorical
73
Relative Frequency Distribution
are frequencies expressed as a fraction (proportion) of the total
74
Cumulative Relative Frequency
The proportion of the total number of observations that fall at or below the upper limit of each group
75
Histograms
is a graphical depiction of a frequency distribution for numerical data
76
Percentiles
Percentiles specify the percent of data entries that are at or below the value of a specific or chosen record
77
Quartiles
Quartiles break the data into four parts
- 25th
- 50th
- 75th
- 100th
78
Cross-Tabulations
- displays the number of observations in a data set for different subcategories of two categorical variables
Ex) gender, edu level, martial status
79
PivotTables
- Used to create cross-tabulations
- Organize and summarize complex data sets into meaningful data
- Drill down into data sets
- Make summary tables and charts
80
Trendlines
are useful for studying the relationship between variables (for example: price vs. demand, bond rate vs. business cycle, intent to purchase product vs. brand attitude, etc.)
81
Types of Trendlines
1. linear
2. polynomial
3. exponential
4. power
82
Regression Analysis
a tool for building mathematical and statistical models that characterize relationships between a dependent (ratio) variable and one or more independent, or explanatory variables (ratio or categorical), all of which are numerical
83
Simple linear regression
involves a single independent variable
84
Multiple regression
involves two or more independent variables
85
Two categories of regression models
1. Cross-sectional data
| 2. Time series data (forecasting)
86
Least Squares Regression
- Goal is to minimize residual error
- The best fit line minimizes the sum of the squares of the residuals
- Errors can be positive or negative, so we square the values to eliminate any negative signs
87
Residuals
are the observed errors associated with estimating the value of the dependent variable using the regression line
88
Regression Significance
The t-statistic will also test the significance of the model coefficients
(Slope and y-intercept)
89
Residual Analysis
- Use the residual plot from the regression tool for analysis
- Residuals in plot should look random, with no clear shape or trend
- Shape or trend means residuals may not be random and could indicate issues with variance, missing terms, outliers, etc.
(Shape => A problem with the regression model)
90
Best Practices Approach
1. Construct a model with all available independent variables. Check for significance of the independent variables by examining the p-values.
2. Identify the independent variable that has the largest p-value that exceeds the level of significance.
3. Remove the single variable identified in step 2 from the model, rerun the model, and evaluate adjusted R2.
4. Repeat steps 2 and 3 until all variables are significant.
91
Forecasting
-Process of predicting a future event
-Underlying basis of all business decisions
(Production, inventory, personnel, facilities)
92
Forecasting Time Horizons
1. Short-range forecast
2. Medium-range forecast
3. Long-range forecast
93
Short-range forecast
- Up to 1 year, generally less than 3 months
| Ex) Purchasing, job scheduling, workforce levels, job assignments, production levels
94
Medium-range forecast
- 3 months to 3 years
| Ex) Sales and production planning, budgeting
95
Long-range forecast
- 3+ years
| Ex) New product planning, facility location, research and development
96
Types of Forecasts
1. Economic forecasts
2. Technological forecasts
3. Demand forecasts
97
Economic forecasts
Address business cycle – inflation rate, money supply, housing starts, etc
98
Technological forecasts
- Predict rate of technological progress
| - Impacts development of new products
99
Demand forecasts
- Predict sales of existing products and services
| - Payoffs in reduced inventory and obsolescence
100
Strategic Importance
- Supply chain management
- Human Resources
- Capacity
101
Supply chain management
Good supplier relations, advantages in product innovation, cost and speed to market
102
Human Resources
Hiring, training, laying off workers
103
Capacity
Shortages can result in undependable delivery, loss of customers, loss of market share
104
7 Forecasting Steps
1. Determine the use of the forecast
2. Select the items to be forecasted
3. Determine the time horizon of the forecast
4. Select the forecasting model(s)
5. Gather the data needed to make the forecast
6. Make the forecast
7. Validate and implement the results
105
Qualitative method of forecasting
- Used when situation is vague and little data exist
- New products, new technology
- Involves intuition, experience
106
Quantitative methods of forecasting
- Used when situation is stable and historical data exist
- Existing products, current technology
- Involves mathematical techniques
107
Types of Qualitative Methods
1. Jury of executive opinion
2. Delphi method
3. Sales force composite
4. Market survey
108
Jury of Executive Opinion
- Pool opinions of high-level experts and managers
- Small group
- Group estimates demand by working together
- Combines managerial experience with statistical models
- Relatively quick
- Disadvantage: Group-think
109
Delphi Method
Panel of experts, queried individually
Iterative process that continues until consensus reached
3 participants: staff, respondents, decision makers
110
Sales Force Composite
- Estimates from individual salespersons are reviewed for reasonableness, then aggregated
- Combined at district and national levels
- Assumed that sales reps know and understand customer wants
- May be overly optimistic
111
Market Survey
- Ask customers about purchasing plans
- Useful for demand and product design and planning
- What consumers say and what they actually do may be different
- May be overly optimistic
112
Quantitative Approaches of forecasting
1. Time-series models
| 2. Associative model
113
Time-series models
- Naïve approach
- Moving averages
- Exponential smoothing
- Trend projection (trendlines)
114
Associative model
Linear regression
115
Naïve approach
- Assumes demand in next period is the same as demand in most recent period
- Can be cost effective and efficient
- good starting point
116
Moving Averages
- Series of arithmetic means
- Subset of historical actual values to generate forecast
- Little or no trend
117
Exponential Smoothing
- Form of weighted moving average
| Most recent data has stronger influence
118
Forecast Accuracy
- We need to obtain the most accurate forecast, no matter the technique
- To find the most accurate forecast, select the model that gives us the lowest forecast error
119
Three measures for error
1. Mean absolute deviation (MAD)
2. Mean squared error (MSE)
3. Mean absolute percent error (MAPE)
120
Mean absolute deviation (MAD)
How much the forecast missed the target
121
Mean squared error (MSE)
Square of how much the forecast missed the target
122
Mean absolute percent error (MAPE)
Average percent error
123
Seasonal Variations in Data
model adjusts trend data for seasonal variations in demand
124
Steps for monthly seasons:
1. Find average historical demand for each month
2. Compute average demand over all months
3. Compute seasonal index for each month
4. Estimate next year’s total demand
5. Divide this estimate of the total demand by the number of months, then multiply by the seasonal index for that month
125
Tracking Signal
- Measures how well the forecast is predicting actual values
| - Ratio of cumulative forecast errors to MAD
