Numerical Issues in Statistical Computing for the Social Scientist

Altman, Micah; Gill, Jeff; McDonald, Michael P.

doi:10.1002/0471475769

Cited by 54 publications

(53 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the last 30 years a large amount of surveys were published demonstrating that different computer programs can produce different results (e.g. Altman et al, 2004;McCullough, 1998McCullough, , 1999McCullough and Wilson, 1999).…”

Section: Introductionmentioning

confidence: 99%

Model-based experimental analysis of enzyme kinetics in aqueous–organic biphasic systems

Zavrel

Schmidt

Michalik

et al. 2007

Journal of Biotechnology

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Model-based experimental analysis of enzyme kinetics in aqueous–organic biphasic systems

Zavrel

Schmidt

Michalik

et al. 2007

Journal of Biotechnology

View full text Add to dashboard Cite

“…The most important options to ensure that a good solution is found, aside from pop.size, are wait.generations, max.generations and hard.generation.limit. Many statistical models have objective functions that are nonlinear functions of the parameters, and optimizing such functions is tricky business (Altman, Gill, and McDonald 2003). Optimization difficulties often arise even for problems that are generally considered to be simple.…”

Section: Resultsmentioning

confidence: 99%

Genetic Optimization Using Derivatives: ThergenoudPackage forR

Mebane¹,

2011

View full text Add to dashboard Cite

Abstractgenoud is an R function that combines evolutionary algorithm methods with a derivativebased (quasi-Newton) method to solve difficult optimization problems. genoud may also be used for optimization problems for which derivatives do not exist. genoud solves problems that are nonlinear or perhaps even discontinuous in the parameters of the function to be optimized. When the function to be optimized (for example, a log-likelihood) is nonlinear in the model's parameters, the function will generally not be globally concave and may have irregularities such as saddlepoints or discontinuities. Optimization methods that rely on derivatives of the objective function may be unable to find any optimum at all. Multiple local optima may exist, so that there is no guarantee that a derivative-based method will converge to the global optimum. On the other hand, algorithms that do not use derivative information (such as pure genetic algorithms) are for many problems needlessly poor at local hill climbing. Most statistical problems are regular in a neighborhood of the solution. Therefore, for some portion of the search space, derivative information is useful. The function supports parallel processing on multiple CPUs on a single machine or a cluster of computers.

show abstract

“…Many techniques such as simulations and nonlinear estimation involve operations carried out an enormous number of times which makes it perfectly possible that the nal result will be accurate to only one or even zero digits if the necessary attention is not given to the computational aspect of solving the econometric problem. For further information about errors in numerical computations, the reader is referred to a source such as Altman et al (2004), which provide a more detailed account on this subject.…”

Section: Sources Of Software Failuresmentioning

confidence: 99%

On the importance of verifying forecasting results

Yalta

Jenal²

2009

International Journal of Forecasting

View full text Add to dashboard Cite

We discuss the various sources of error in numerical computations with the use of examples from the literature relevant to time series analysis. We also submit a case where, by manual verication, we were able to discover a plausible forecast to be erroneous due to a number of software aws in the XLSTAT addin for Microsoft Excel. Furthermore, after discussing the alternative techniques for implementing on a computer the ARIMA (AutoRegressive, Integrated, Moving Average) methodology, we show that dierent approaches can cause considerable discrepancies in the results across dierent programs and even within a single software system.

show abstract

Numerical Issues in Statistical Computing for the Social Scientist

Cited by 54 publications

References 0 publications

Model-based experimental analysis of enzyme kinetics in aqueous–organic biphasic systems

Model-based experimental analysis of enzyme kinetics in aqueous–organic biphasic systems

Genetic Optimization Using Derivatives: ThergenoudPackage forR

On the importance of verifying forecasting results

Contact Info

Product

Resources

About