Accuracy of error propagation exemplified with ratios of random variables

Abstract

A quantum central limit theorem for sums of independent identically distributed random variables On the statistical errors in the estimate of acoustical energy density by using two microphones in a one dimensional fieldThe method of error propagation provides a convenient tool for calculating mean and variance of a measurand from means and variances of primarily measured quantities. However, being based on a ͑usually first-order͒ Taylor approximation of the measurement function, it only yields approximate results with unknown accuracy. We develop a method for estimating the accuracy of ͑Nth-order͒ error propagation for an arbitrary number of correlated random quantities, and apply our findings to the ratio of two random variables ͑RVs͒. A comparison with some analytically solved expressions for certain probability density functions ͑PDFs͒ as well as with some computer simulations reveals the excellent quality of our estimates as long as the involved PDFs are not significantly skew. For the ratio of two RVs it turns out that conventional, first-order error propagation is safely applicable ͑with about 1% accuracy͒ as long as the denominator's mean is larger than about 12 times its standard deviation. Using second-order error propagation, the approximation for the ratio's mean can be refined, yielding 1% accuracy if the denominator's mean is larger than about four times its standard deviation.

Estimations of total ecosystem carbon pools distribution and carbon biomass current annual increment of a moist tropical forest

“…To account for error associated at different levels (regression equations, nested plots, different land uses and strata) we used the propagation error method which have been proved by several studies to be appropriate for estimating uncertainties of a parameter resulting from different variables associated each with different errors (Winzer, 2000;Chave et al, 2001Chave et al, , 2004. The model used is:…”

Reach and effectiveness of mailed nicotine replacement therapy for smokers: 6-month outcomes in a naturalistic exploratory study

“…Variance estimates for the OTS arm were appropriate to the stratified sampling design and weighting using Taylor-Series variance estimation. Wald CIs for RR and RD estimates were obtained using the first Taylor-series expansion using variance estimates calculated separately for each arm 25 26…”

Modelling mean above and below ground litter production based on yield tables

“…An alternative approach would have been error propagation methods. However, error propagation methods assume specific distributions of the parameters and all intermediate results, and they have problems with non-linearity, because they are usually based on a first order Taylor expansion (Winzer 2000). Contrary, Monte Carlo analysis is an effective method to assess uncertainty when models are complex with non-linearity and with different types of correlations (Morgan andHenrion 1990, Vose 1996).…”