A Generalized Rounding Alternative
نویسنده
چکیده
It is important to understand the consequences of using the ROUND function in the SAS System. By default, numbers with a trailing 5 are rounded away from zero (i.e., 0.15 to 0.2 and -0.25 to -0.3), neglecting any numerical precision errors that a computer’s hardware limitations may introduce. In some instances, this type of rounding can lead to an overestimation of the true mean of a set of numbers, as illustrated by the simulation described in this paper. The macro code presented in this paper eliminates this bias by randomly rounding up or down, depending on the outcome of a normally distributed random variable.
منابع مشابه
Cox model versus generalized logrank test for time–to–event data with ties
The Cox proportional hazards model is the standard tool for relative risk estimation and inference for survival studies. The generalized logrank (GLR) approach (Mehrotra and Roth, 2001) has been proposed as a more efficient alternative to the Cox model when the number of subjects is small and there are no ties in the data. However, ties will result if continuous data are grouped by rounding to ...
متن کاملThe Equality Constrained Indefinite Least Squares Problem : Theory and Algorithms ∗ †
We present theory and algorithms for the equality constrained indefinite least squares problem, which requires minimization of an indefinite quadratic form subject to a linear equality constraint. A generalized hyperbolic QR factorization is introduced and used in the derivation of perturbation bounds and to construct a numerical method. An alternative method is obtained by employing a generali...
متن کاملMachine-checkable Correctness Proofs
This talk will present an effort to formalize Taylor Models in the Coq proof assistant. Machinecheckable correctness proofs are facilitated by an abstract viewpoint: Taylor models can be generalized to balls in the Chebyshev metric. Extensions of elementary functions are then explained as compositions of such balls. This approach also accommodates other polynomial approximation methods than Tay...
متن کاملA New Approximation Technique for Resource-Allocation Problems
We develop a rounding method based on random walks in polytopes, which leads to improved approximation algorithms and integrality gaps for several assignment problems that arise in resource allocation and scheduling. In particular, it generalizes the work of Shmoys & Tardos on the generalized assignment problem in two different directions, where the machines have hard capacities, and where some...
متن کاملFast Enclosure for Solutions of Generalized Least Squares Problems∗
Fast algorithms for enclosing solutions of generalized least squares problems are proposed. To develop these algorithms, theories for obtaining error bounds on numerical solutions are established. The error bounds obtained by these algorithms are “verified” in the sense that all the possible rounding errors have been taken into account. A technique for obtaining smaller error bounds is introduc...
متن کامل