10.06.2016 Views

eldo_user

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Monte Carlo Analysis<br />

How to Specify a Distribution Function<br />

.DISTRIB TRI (0, 0) (0.66666, 1) (1, 0)<br />

.PARAM XVAL=5 LOT/TRI=15<br />

V1 1 0 XVAL<br />

.OP<br />

.MC 1000 NBBINS=50<br />

.EXTRACT DC LABEL=R1 E(V1,V)<br />

.END<br />

The distribution parameters a, b are defined implicitly by construction. The most frequent value<br />

m is obtained by the formula: 5 × 0.66666 + 5 ≈ 15. Note also that the value XVAL=5 in the<br />

PARAM statement cannot be considered as the nominal value of the resistor. It is used as a shift<br />

in value of XVAL and LOT/TRI=15 is a scaling factor.<br />

Other Common Distributions<br />

You can generate additional distributions by using the basic building blocks of the previous<br />

section.<br />

• Half-Normal Distribution<br />

The half-normal distribution is the probability of the absolute value of a normal<br />

distribution with mean 0 and variance σ 2 . If X ~ N(0, σ 2 ), then Y = |X| is half-normally<br />

distributed.<br />

• Log-Normal Distribution<br />

The log-normal distribution is the single-tailed probability distribution of any random<br />

variable whose logarithm is normally distributed. If X is a random variable with a<br />

normal distribution, then Y = e X has a log-normal distribution.<br />

Related Topics<br />

Tip<br />

See “.DISTRIB” in the Eldo Reference Manual.<br />

Correlation Between Gaussian Input Variables<br />

Specifying a Monte Carlo Analysis<br />

436<br />

Eldo® User's Manual, 15.3

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!