We investigate the setting in which Monte Carlo methods are used and draw a parallel to the formal setting of statistical inference. In particular, we nd that Monte Carlo approximation gives rise to a bias-variance dilemma. We show that it is possible to construct a biased approximation schemes with a lower approximation error than a related unbiased algorithms.

The norm of practice in estimating graph properties is to use uniform random node (RN) samples whenever possible. Many graphs are large and scale-free, inducing large degree variance and estimator variance. This paper shows that random edge (RE) sampling and the corresponding harmonic mean estimator for average degree can reduce the estimation variance significantly. First, we demonstrate that ...

Inference using di erence-in-di erences with clustered data requires care. Previous research has shown that, when there are few treated clusters, t tests based on a clusterrobust variance estimator (CRVE) severely over-reject, di erent variants of the wild cluster bootstrap can over-reject or under-reject dramatically, and procedures based on randomization inference show promise. We demonstrate...

Automatic speech recognizers perform poorly when training and test data are systematically different in terms of noise and channel characteristics. One manifestation of such differences is variations in the probability density functions (pdfs) between training and test features. Consequently, both automatic speech recognition and automatic speaker identification may be severely degraded. Previo...