WWW | Store | Account | Contact Us
SHAZAM Homepage

FUZZY

The FUZZY command facilitates the construction of fuzzy-set econometric models. Fuzzy set econometric modelling provides a means of using the information associated with economic concepts and linguistic economic variables to predict the value of another economic variable.

In general, the format is:

FUZZY x1 x2 / rules=rulemat degrees=degreemat options

where rulemat and degreemat are 5 x 5 matrices of decision rules and degrees.

The available options are:

OPTION DESCRIPTION
beg= Specifies the BEGinning observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect.
cma= Specifies the number of periods to start the Cumulative Moving Average. If this is used CMA-1 observations will be lost.
degrees= Specifies a 5 x 5 matrix of degrees of association to be used with the decision rules. The 25 degrees of association are represented in a (5x5) matrix. A row of this matrix corresponds to a membership level for the first causal variable. A column corresponds to a membership level for the second causal variable. This option is required.
dump DUMPS large amounts of output mainly useful to SHAZAM Consultants
end= Specifies the ENDing observation to be used in estimation. This option overrides the SAMPLE command and defaults to the sample range in effect.
graphdata Provides a graphics plot of the data.
graphrule Provides a graphics plot of the rule surface.
median Specifies to center the data around the median instead of the mean.
nolist Suppresses the List of computed index values and rules.
nopmatrix Suppresses printing of the rules and degrees matrices.
nostandard Assumes that the input data is standardized.
passoc Prints the table of Association values.
pbreak Prints the table of Breakpoints.
predict= Saves the calculated index in the variable specified.
rma= Specifies the number of periods for the Regular Moving Average. RMA-1 observations will be lost
rules= Specifies a 5 x 5 matrix of decision rules. This option is required.
weight= Specifies a 5 element vector of weights to replace the default of (0, 0.25, 0.5, 0.75, 1.0).