System Variables

Revision as of 19:31, 4 February 2015 by Lchrisman (talk | contribs)

<<prev | next>>

System variables (abbreviated as sysvars) are pre-defined variables in Analytica. You can modify system variables to control the way things are printed, plotted etc. Many system variables control the formatting of graphs. You can also include system variables in the definitions of user-specified variables. You can use the Show command to print a system variable. You may change the definition of a system variable with the usual syntax for definition assignment, e.g.,

 Example>checking:0

If the definition is outside of the legal range for that variable, Analytica will issue the following message:

 Example>checking:4
 Value for Checking must be an integer between 0 and 1 inclusive.
 Do you want to edit the Definition of Checking?

System variables for controlling the format of graphs and other plots are listed in the Summary section of the chapter on graphing.

Abbreviation
System variable determining Analytica' ability to understand abbreviations. abbr: 1 means abbreviations are accepted, abbr: 0 means abbreviations are not permitted.


Checking
System variable telling Analytica to examine all of the Check attributes in a model and to flag the first incidence of a definition that is in conflict with its check. checking:1 switches it on (default); checking:0 switches it off

Heapsize
System variable, printed out when you use the Clock command. The amount of memory in use by Analytica. Normal_fracs System variable used in calculating normal distributions.

Numberwidth
A System variable that controls the width used in printing out numbers. Default is 4. Values less than 4 aren't recommended. If Numberwidth is set to 12 or more, numbers will be printed in floating point "E" format (e.g., 1.234456789E+12). The value 0 means "free format", and Analytica will choose what width it likes. Syntax: numberw: x

Run
System variable that indexes a sample distribution.

Samplesize
System variable that determines the number of elements in a sample distribution. samplesize:100 is the default; samplesize: 32000 the maximum in the Professional and Free 101 editions, limited by memory in Enterprise edition and up.

SampleType
System variable specifying the procedure for sampling a distribution, i.e., Median Latin Hypercube (Sampletype: 0), Random Latin Hypercube (Sampletype: 1), or Simple Monte Carlo (SampleType: 2). In Analytica 4.3 there are also two values: biased MLH (sampleType:3) and biased RLH (sampleType:4) that emulate the latin hypercube shuffle used by Analytica 4.2 and earlier. You only want to use those if you need to reproduce the exact sample computed by Analytica 4.2. (Analytica 4.3 removes a small bias in the shuffling algorithm, hence you get slightly different Latin Hypercube samples)

Time
System variable, usually used as an index for Dynamic. For dynamic models, this must be assigned a list of the time points (e.g. years) at which the Dynamic variables are to be evaluated.

Verbosity
System variable. Controls how "chatty" Analytica is. verb: 1 verbose; verb: 2 does not print tables unless explicitly requested; verb: 4 prints "evaluating " when it is; verb: 8 is Filetrace; verb: 16 is debug mode. verb:128 (128 is new to Analytica 4.3) causes the evaluating trace to be indented. (Verbosity:134 is useful for debugging dynamic loop cycles).

<<prev | next>>

Comments


You are not allowed to post comments.