The Model Analysis Report provides a summary of the business
element regression analysis. This report is used to evaluate
whether the model acceptably represents the application.
Although the values of r-squared, F, and p indicate the
statistical significance of the proposed model, you should
inspect the intercept value and coefficients to see if the
model is reasonable.
A sample Model Analysis Report is shown in Figure 10-1. The
report includes the following information:
MODEL OF: Dependent data element named as the element on
the Business Element Forecasting control screen
(shown in Figure 10-3).
BASED ON: Business elements proposed by the analyst for
the model.
The first four columns of the report are discussed below.
R**2: The r-squared value for the model. Although the
SAS REG stepwise option procedure attempts to
use every one of the business elements that you
suggest, Business Element Forecasting excludes
any term that does not result in a minimum
improvement in the r-squared value (default
0.05, which may be altered by the DELRSQ control
statement). For this example, each subsequent
r-squared represents at least an increase of
0.05. Hence, all of the terms are included in
the model developed by the program. We
recommend an r-squared value of 0.70 and above
for the acceptance of models produced using this
software.
F: The F statistic for the model. The F
statistic is calculated from the ratio of sums
of squares of the model. Larger F values
indicate a more reliable model. Note that the F
statistics may be unreliable for models
developed with less than 30 historical data
points. An F value is provided for each step
made by the regression procedure.
P: The probability that the model proposed is
significant, that is, more reliable than
forecasting the average value of the historical
observations. Although most statistics texts
recommend testing the null hypothesis to a 0.001
level, 0.01 is recommended for models based on
computer measurement data. A p value is
provided for each step made by the regression
procedure.
INTERCEPT: The y intercept (b value) for the regression
equation. The intercept value should be smaller
than the actual historical data observations.
Otherwise, a model has been developed that
indicates that resource consumption decreases
with increases to input.
The next four columns are produced for each business element
you specify as independent elements on the Business Element
Forecasting control screen (Figure 10-3). In this example,
the business elements are invoices, updates, and open items.
(These four columns appear under the heading BUSINESS
ELEMENTS.) These columns contain the following:
NAME: The name of the business element being analyzed.
COEFFICIENT: The coefficient for the business element in
the regression equation. The coefficient is the
m value in the regression equation. Negative
coefficients indicate that you can expect
resource consumption to decrease with increases
in the business element value.
F: The F statistics calculated for the business
element. This F value indicates the
significance of the business element in the
model that was developed. Once again, larger
values indicate a more reliable model and the
value may be misleading for models based on
small samples.
P: The probability that the specified business
element makes a significant contribution to the
model developed. We recommend a 0.01 acceptance
value.
CA MICS Capacity Planner ANALYSIS OF BUSINESS ELEMENT MODELS MODEL OF: BILEXCPS BASED ON: INVOICES UPDATES OPN_ITEM ----------BUSINESS-ELEMENTS---------- R**2 F P INTERCEPT NAME COEFFICIENT F P -------- ------- ----- --------- -------- ----------- ------- ----- 0.78 17.26 .0089 7768.21 INVOICES -0.159782 17.26 .0089 0.91 20.86 .0077 9091.08 INVOICES -0.14591 28.35 .0060 OPN_ITEM -.0328758 6.27 .0665 0.98 44.36 .0055 7938.98 INVOICES -0.122026 47.07 .0063 UPDATES 0.196209 8.91 .0584 OPN_ITEM -.0309463 16.42 .0271
Figure 10-1. Model Analysis Report
|
Copyright © 2014 CA.
All rights reserved.
|
|