Page 241 - qrps

Basic HTML Version

© 2014 Pacific Crest
241
Technique
L
ookup
T
able
Category
Ease of Use
Modifying
Description
The replacement of values in an attribute with new values based upon conditional
rules
Benefits
Limitations
Can be used for clustering data, recording, building
frequencies, and remapping
Data not fitting a rule will become missing values
Tool
Lookup
Application
Survey data, restructuring relationships, organizing relationships
Example
Lookup(“cities”,”city/States”) Produce a state mapping
Technique
M
erge
Category
Ease of Use
Combining
Description
Integrating the data from two data sets into a single data file
Benefits
Limitations
Can relate two sets of data based upon the key
attribute that is being studied
The key attribute occurrence must be in each data file
or that value will be missing associated data
Tool
Mapping function or Merging function
Application
Comparison studies - like high school vs. college performance
Example
Merge(“High School”, “College”,StudentID)
Technique
M
oving
A
vg
Category
Ease of Use
New Attribute
Description
Find the average of a number of periods both before and after the current period
Benefits
Limitations
Smoothing out trend data so the trend data can be
seen more clearly
There will be missing values at the beginning and end
of the data because of no data to average
Tool
Moving avg or a combination of functional transformations and Lag variables
Application
Time series analysis - Reducing variability to see patterns
Example
Movingavg(“Price”, 5) will produce a 5 day moving average of
Price
5.4 Transforming Data