Comparison of the Akaike Information Criterion, the Schwarz criterion and the F test as guides to model selection

Ludden, Thomas M.; Beal, Stuart L.; Sheiner, Lewis B.

doi:10.1007/bf02353864

Cited by 270 publications

(179 citation statements)

References 13 publications

Supporting

Mentioning

172

Contrasting

Unclassified

Order By: Relevance

“…where RSS is the sum of squared residuals for the model being tested (Burnham & Anderson, 2004;Ludden, Beal, & Sheiner, 1994). (The full calculation of BIC includes an additional parameter that estimates the population variance from the sample, but this parameter has been subtracted out of the formula used here because, as an added constant, it should be irrelevant to any comparison between different BIC scores based on the same data.)…”

Section: Resultsmentioning

confidence: 99%

The acquisition of conditioned responding.

Harris

2011

Journal of Experimental Psychology: Animal Behavior Processes

View full text Add to dashboard Cite

This report analyzes the acquisition of conditioned responses in rats trained in a magazine approach paradigm. Following the suggestion by Gallistel, Fairhurst and Balsam (2004), Weibull functions were fitted to the trial-by-trial response rates of individual rats. These showed that the emergence of responding was often delayed, after which the response rate would increase relatively gradually across trials. The fit of the Weibull function to the behavioral data of each rat was equaled by that of a cumulative exponential function incorporating a response threshold. Thus the growth in conditioning strength on each trial can be modeled by the derivative of the exponential -a difference term of the form used in many models of associative learning (e.g., Rescorla & Wagner, 1972). Further analyses, comparing the acquisition of responding to a continuously reinforced stimulus (CRf) and a partially reinforced stimulus (PRf), provided further evidence in support of the difference term. In conclusion, the results are consistent with conventional models that describe learning as the growth of associative strength, incremented on each trial by an error-correction process.

show abstract

Section: Resultsmentioning

confidence: 99%

The acquisition of conditioned responding.

Harris

2011

Journal of Experimental Psychology: Animal Behavior Processes

View full text Add to dashboard Cite

show abstract

“…Generalized additive modeling (GAM), implemented in the Xpose library of 'R' (ver. 2.11.1; R Foundation for Statistical Computing, Vienna, Austria [10]), was used for numerical screening. Only the variables showing a positive result in this screening and having physiological relevance to PK parameters were included in the model and evaluated using LRT.…”

Section: Covariate Selectionmentioning

confidence: 99%

Population pharmacokinetic analysis of the multiple peaks phenomenon in sumatriptan

Lee

Lim

Seong

et al. 2015

Transl Clin Pharmacol

View full text Add to dashboard Cite

+The first two authors contributed equally to this work.The objective of this study was to develop a population pharmacokinetic (PK) model for sumatriptan, which frequently shows an atypical absorption profile with multiple peaks. Sumatriptan, a selective agonist for the vascular serotonin (5-HT1) receptor that causes vasoconstriction of the cerebral arteries, is used for the acute treatment of migraine attack with or without aura. Despite its relatively high between-subject variability, few reports have addressed PK modeling of sumatriptan. Plasma data obtained after a single 50-mg oral dose of sumatriptan in 26 healthy Korean male subjects were used. Blood samples were collected 0 (predose), 0.25, 0.5, 0.75, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, and 12 h after dosing. Plasma sumatriptan concentrations were analyzed using UPLC/MS/MS. Population PK analysis was performed using plasma concentration data for sumatriptan with NON-MEM (ver. 7.2). A total of 364 concentrations of sumatriptan were captured by a one-compartment model with first-order elimination, and a combined transit compartment model and first-order absorption with lag time was successful in describing the PK with multiple peaks in the absorption phase of sumatriptan. The creatinine clearance as a covariate significantly (P < 0.01) influenced the absorption fraction (f ). The final model was validated through a visual predictive check and bootstrapping with no serious model misspecification.

show abstract

“…For two nested models a decrease in 3.84 or 6.63 points in -2LL for an extra parameter was considered significant at the 5 or 1% level, respectively. Since some of the models compared were not nested, -2LL was not used directly for comparative purposes, and the value of the Akaike information criteria (AIC) [22] computed as -2LL ? 2 9 Np, where Np is the number of the parameters in the model, was used instead.…”

Section: Model Selectionmentioning

confidence: 99%

Semi-mechanistic population pharmacokinetic/pharmacodynamic model for neutropenia following therapy with the Plk-1 inhibitor BI 2536 and its application in clinical development

Soto

Staab

Tillmann

et al. 2010

Cancer Chemother Pharmacol

View full text Add to dashboard Cite

Purpose (1) To describe the neutropenic response of BI 2536 a polo-like kinase 1 inhibitor in patients with cancer using a semi-mechanistic model. (2) To explore by simulations (a) the neutropenic effects for the maximum tolerated dose (MTD) and the dose at which dose-limiting toxicity occurred, (b) the possibility to reduce the cycle duration without increasing neutropenia substantially, and (c) the impact of the initial absolute neutrophil count (ANC) on the degree of neutropenia for different doses. Experimental design BI 2536 was administered as intravenous infusion over 60 min in the dose range from 25 to 250 mg. Three different administration schedules were explored: (a) day 1, (b) days 1, 2, and 3 or (c) days 1 and 8 within a 3 week treatment cycle. BI 2536 plasma concentrations and ANC obtained during the first treatment cycle from 104 patients were analysed using the population approach with NONMEM VI. Results Neutropenia was described by a semi-mechanistic model resembling proliferation at the stem cell compartment, maturation, degradation, and homeostatic regulation. BI 2536 acts decreasing proliferation rate.Simulations showed that (1) all MTD doses showed an acceptable risk of neutropenia, (2) when BI 2536 is given as 200 mg single administration, cycle duration can be reduced from 3 to 2 weeks, and (3) baseline ANC might be considered to individualise the dose of BI 2536. Conclusions A semi-mechanistic population model was applied to describe the neutropenic effects of BI 2536. The model was used for simulations to support further clinical development.

show abstract

Comparison of the Akaike Information Criterion, the Schwarz criterion and the F test as guides to model selection

Cited by 270 publications

References 13 publications

The acquisition of conditioned responding.

The acquisition of conditioned responding.

Population pharmacokinetic analysis of the multiple peaks phenomenon in sumatriptan

Semi-mechanistic population pharmacokinetic/pharmacodynamic model for neutropenia following therapy with the Plk-1 inhibitor BI 2536 and its application in clinical development

Contact Info

Product

Resources

About