7. UNIVARIATE MODEL FORECASTING › 7.6 Case Studies › 7.6.1 Extended Example Univariate Model Forecasting › 7.6.1.3 Output Reports › 7.6.1.3.1 Model Analysis Report

7.6.1.3.1 Model Analysis Report

Figure 7-17 shows the Model Analysis Report for the channel 5
I/O data.  Note that the value of r-squared is a reasonable
0.75.  The F value of 30.84 appears to be large and the
p-value associated with this F is 0.0004, which is very
small.  You can interpret this p-value as the probability
that you would have estimated this model, given that the data
was purely random, and that no linear relationship truly
exists.  The p-value says that there are no more than four
chances in 10,000 that these estimates would arise due to
random error.

The model that is shown in the report is in the following
form:

    CHAN5CNT = 9736797 + 587938 * T1

In this model, T1 is the number of the month and CHAN5CNT is
the monthly total I/O count for channel 5.  This model
indicates that you can expect I/O loads to rise by about
588,000 I/Os each month.
 

                                              CA MICS Capacity Planner

                                        ANALYSIS OF UNIVARIATE FORECASTING MODEL

                    MODEL OF: CHAN5CNT
                    BASED ON: LINEAR
                                                             ------------TIME ELEMENTS------------
                      R**2        F        P     INTERCEPT     NAME    COEFFICIENT     T       P
                    --------   -------   -----   ---------   --------  -----------  -------  -----
                      0.75      30.84    .0004    9736797    LINEAR        587938     5.55   .0004

 
Figure 7-17.  Extended Example Model Analysis Report