EBTM 350 FINAL EXAM VERIFIED STUDY GUIDE
Creating good decision models requires: - Answers - -Solid understanding of business
functional areas
-Knowledge of business practice and research
- Logical Skills
empirical data - Answers - Information gathered from sensory observation and
experimentation
-Observe something; then develop model that best reflects what we observed
spreadsheet engineering - Answers - The process of developing good, useful, and
correct spreadsheet models
Spreadsheet models characterize: - Answers - The relationship between inputs and
outputs
Do not use input data in model formulas instead: - Answers - reference the spreadsheet
cells that contain the data
Verification - Answers - the process of ensuring that a model is accurate and free from
logical errors
3 approaches to spreadsheet engineering that improve quality - Answers - 1. Improve
the design and format of the spreadsheet
2. Improve the process used to develop the spreadsheet
3. Inspect your results carefully and use appropriate tools available in Excel
Improve the design and format of the spreadsheet - Answers - -Clearly define input and
output areas
-Break complex formulas into small pieces
-Make the spreadsheet end-user friendly
Improve the process used to develop the spreadsheet - Answers - -Be careful when
copying and editing formulas
-Check results as build sections of the spreadsheet
Inspect your results carefully and use appropriate tools available in Excel - Answers - -
Formula auditing tools in formula tab on ribbon
-Lock cells, check work, etc.
Validity - Answers - the degree to which a model represents reality
, Newsvendor Problem - Answers - problem where a one-time (Single Period) decision is
made under uncertain demand.
Ex: Seasonal retail inventory, game programs, dated material
Parameterize - Answers - To give a model parameters.
Data Mining - Answers - The nontrivial process of identifying valid, novel, potentially
useful, and ultimately understandable patterns in data stored in structured databases.
Patterns in data mining - Answers - explanatory- explains interrelationships and
affinities among the attributes
Predictive- foretelling future values of certain attributes
Process - Answers - Data mining comprises many iterative steps
Nontrivial - Answers - some experimentation is involved- not straightforward
computation of predefined quantities
Valid - Answers - Discovered patterns should hold true on new data(with reasonable
certainty)
Novel - Answers - Patterns not previously known to user
potentially useful - Answers - Discovered patterns should usually lead to some benefit to
a task / user
ultimately understandable - Answers - patterns should make business sense and be
intuitive
Two approaches to data mining - Answers - -Supervised Method
-Unsupervised Method
Supervised Method - Answers - algorithm is " trained" by some of the data; some
descriptive attributes are defined computer is asked to find the rest(EX:Regression
Analysis)
Unsupervised Method - Answers - Computer asked to both define and find relationships
;Training data only contains descriptive attributes no model or hypothesis prior to
analysis(EX:Cluster analysis)
Methods within supervising & unsupervising - Answers - 1. Prediction
2. Association
3. Cluster
Prediction - Answers - forecasting the future using a label or number
Creating good decision models requires: - Answers - -Solid understanding of business
functional areas
-Knowledge of business practice and research
- Logical Skills
empirical data - Answers - Information gathered from sensory observation and
experimentation
-Observe something; then develop model that best reflects what we observed
spreadsheet engineering - Answers - The process of developing good, useful, and
correct spreadsheet models
Spreadsheet models characterize: - Answers - The relationship between inputs and
outputs
Do not use input data in model formulas instead: - Answers - reference the spreadsheet
cells that contain the data
Verification - Answers - the process of ensuring that a model is accurate and free from
logical errors
3 approaches to spreadsheet engineering that improve quality - Answers - 1. Improve
the design and format of the spreadsheet
2. Improve the process used to develop the spreadsheet
3. Inspect your results carefully and use appropriate tools available in Excel
Improve the design and format of the spreadsheet - Answers - -Clearly define input and
output areas
-Break complex formulas into small pieces
-Make the spreadsheet end-user friendly
Improve the process used to develop the spreadsheet - Answers - -Be careful when
copying and editing formulas
-Check results as build sections of the spreadsheet
Inspect your results carefully and use appropriate tools available in Excel - Answers - -
Formula auditing tools in formula tab on ribbon
-Lock cells, check work, etc.
Validity - Answers - the degree to which a model represents reality
, Newsvendor Problem - Answers - problem where a one-time (Single Period) decision is
made under uncertain demand.
Ex: Seasonal retail inventory, game programs, dated material
Parameterize - Answers - To give a model parameters.
Data Mining - Answers - The nontrivial process of identifying valid, novel, potentially
useful, and ultimately understandable patterns in data stored in structured databases.
Patterns in data mining - Answers - explanatory- explains interrelationships and
affinities among the attributes
Predictive- foretelling future values of certain attributes
Process - Answers - Data mining comprises many iterative steps
Nontrivial - Answers - some experimentation is involved- not straightforward
computation of predefined quantities
Valid - Answers - Discovered patterns should hold true on new data(with reasonable
certainty)
Novel - Answers - Patterns not previously known to user
potentially useful - Answers - Discovered patterns should usually lead to some benefit to
a task / user
ultimately understandable - Answers - patterns should make business sense and be
intuitive
Two approaches to data mining - Answers - -Supervised Method
-Unsupervised Method
Supervised Method - Answers - algorithm is " trained" by some of the data; some
descriptive attributes are defined computer is asked to find the rest(EX:Regression
Analysis)
Unsupervised Method - Answers - Computer asked to both define and find relationships
;Training data only contains descriptive attributes no model or hypothesis prior to
analysis(EX:Cluster analysis)
Methods within supervising & unsupervising - Answers - 1. Prediction
2. Association
3. Cluster
Prediction - Answers - forecasting the future using a label or number
