Search optimal lambda
[lambdaopt,loglikeopt] = boxcoxestimate(data) [lambdaopt,loglikeopt] = boxcoxestimate(data,lambda0)
a m-by-1 matrix of doubles, greater or equal to zero, the data to transform
a double, the initial value of lambda (default = 0)
a double, the optimal lambda
a double, the optimal value of the log-likelihood
boxcoxestimate maximizes the log-likelihood of the Box-Cox transformation using a 1D optimization. The value of lambda which maximizes the log-likelihood is returned. The log-likelihood is evaluated by the boxcoxloglikelihood function.