For an IID sample from with , a nonconstant observed sample has maximum-likelihood estimatorsCompleting the square first minimizes the residual sum at ; differentiating the profiled log-likelihood in gives the second formula. It uses denominator , whereas the unbiased sample variance uses : its expectation is . A constant sample, including any one-observation sample, has an unbounded likelihood as the variance decreases to zero, so the displayed zero variance is only a boundary extension in that case.
Articles by others on the same topic
There are currently no matching articles.