Sabre |
ARITHMETIC | ||
Sabre manual |
Syntax: ARITHMETIC option where option is FAST or ACCURATE.
Purpose: If option is FAST, then real numbers are stored in standard floating point format, which may potentially give rise to underflow problems for long sequences of observations for the same case. If option is ACCURATE, then real numbers are stored in mantissa-exponent format, which prevents underflow but may significantly decrease the speed of the fitting algorithm. If underflow does occur while using the fast method, an error message will be returned and the fitting abandoned. The user can then issue the ARITHMETIC ACCURATE command and refit the model. Note that option may be abbreviated to F or A. The default arithmetic type is FAST. The current setting of the ARITHMETIC parameter may be obtained by issuing the DISPLAY SETTINGS command. The ARITHMETIC command has no effect on model fitting instigated using the LFIT command. |
Sabre manual |
Other links: Centre for e-Science | Centre for Applied Statistics