Truncated Lognormal Estimation (Scripts) Publisher's description
from Mike Sheppard
[MU, SIGMA, D] = TRUNCLOGNORMEST(DATA) returns the estimated parameters of the lognormal distribution
[MU, SIGMA, D] = TRUNCLOGNORMEST(DATA) returns the estimated parameters of the lognormal distribution, with unknown left truncation, using the sample data of percent of observations greater than observed quantities.
DATA is assumed to be nx2 where the first column is a set of known values, and the second column is the fraction of observations greater than those values
Solves for the parameters [mu, sigma, d] such that P(X>x|X>d) for lognormal distribution with parameters (mu, sigma) that fits the data with minimal residual error.
data=[ 7.9500 0.8835; 8.1273 0.8627; 8.2057 0.8536; 23.0609 0.1068; 25.7043 0.0759];
TruncLogNormEst(data) gives [2.2077 0.6325 6.9900]
Actual parameters were [2.1972 0.6366 7.0000]
In this example, observations less than 7 could not be recorded - due to instrument limitations or other reasons. Of those observations that were recorded, 10.68% were greater than 23.0609. Without truncation the observations would follow a lognormal distribution with parameters [2.2077,0.6325]. Using the sample data the estimates were calculated of those unknown parameters.
System Requirements:MATLAB 7.9 (2009b)
Program Release Status: New Release
Program Install Support: Install and Uninstall