Time Series Analysis Package
DPF Object
The DPF class object functions take the output of the FREQ class object as input. The DPF class object performs count distribution analysis for a time series.
Table 13 summarizes the methods that are associated with the DPF object.
Table 13: Methods of the DPF Object
| Method | Description |
|---|---|
| Initialize | Initialize the DPF object |
| Run | Run the analysis |
| SetOption | Set the specifications |
Figure 8 diagrams the methods of the DPF object.
Figure 8: DPF Data Flow
DPF Synopsis
DECLARE OBJECT obj (DPF);
Method syntax, in order of typical usage:
rc=obj.Initialize (FREQObj );
rc=obj.SetOption ('Name',Value);
rc=obj.Run ();
DPF Methods
DPF.Initialize Method
rc=obj.Initialize (FREQObj );
Specifies a FREQ object, which includes frequency analysis results. This method must be called before the time series arrays and other attributes are specified for the DPF object.
Input Arguments
You must specify a FREQ object, which includes frequency analysis results.
- FREQObj
specifies the FREQ object as input.
DPF.Run Method
rc=obj.Run ();
Runs the DPF object to perform the count distribution analysis. Upon successful completion, various results can be extracted from the DPF object.
Input Arguments
You can specify one of the following for 'Name' and its associated Value:
- ALPHA
specifies the confidence level size, where alpha must be between 0 and 1. The default value is 0.05.
- CONVERGE
specifies the convergence criterion.
- MAXITER
specifies the maximum number of iterations, where Value is an integer.
- 'SELECT'
-
takes a string Value that specifies the distribution selection criterion. You can specify one of the following Values:
- AIC
specifies Akaike’s information criterion.
- BIC
specifies the Bayesian information criterion.
- LOGLIK
specifies the log-likelihood.
The default value is LOGLIK.
- 'METHOD'
-
takes a string Value that specifies the candidate distribution to use in the analysis. You can specify one of the following Values:
- BEST
specifies the best distribution, based on the value of the select argument.
- BINOMIAL
specifies the binomial distribution.
- GEOMETRIC
specifies the geometric distribution.
- NEGBINOMIAL
specifies the negative binomial distribution.
- POISSON
specifies the Poisson distribution.
- ZMBINOMIAL
specifies the zero-modified binomial distribution.
- ZMGEOMETRIC
specifies the zero-modified geometric distribution.
- ZMNEGBINOMIAL
specifies the zero-modified negative binomial distribution.
- ZMPOISSON
specifies the zero-modified Poisson distribution.
The default distribution is BEST.