Matches in SemOpenAlex for { <https://semopenalex.org/work/W2018195004> ?p ?o ?g. }
- W2018195004 endingPage "483" @default.
- W2018195004 startingPage "475" @default.
- W2018195004 abstract "The performance of a prediction model is usually worse in external validation data compared to the development data. We aimed to determine at which effective sample sizes (i.e., number of events) relevant differences in model performance can be detected with adequate power.We used a logistic regression model to predict the probability that residual masses of patients treated for metastatic testicular cancer contained only benign tissue. We performed standard power calculations and Monte Carlo simulations to estimate the numbers of events that are required to detect several types of model invalidity with 80% power at the 5% significance level.A validation sample with 111 events was required to detect that a model predicted too high probabilities, when predictions were on average 1.5 times too high on the odds scale. A decrease in discriminative ability of the model, indicated by a decrease in the c-statistic from 0.83 to 0.73, required 81 to 106 events, depending on the specific scenario.We suggest a minimum of 100 events and 100 nonevents for external validation samples. Specific hypotheses may, however, require substantially higher effective sample sizes to obtain adequate power." @default.
- W2018195004 created "2016-06-24" @default.
- W2018195004 creator A5009766171 @default.
- W2018195004 creator A5035355889 @default.
- W2018195004 creator A5056077573 @default.
- W2018195004 creator A5089919157 @default.
- W2018195004 date "2005-05-01" @default.
- W2018195004 modified "2023-10-11" @default.
- W2018195004 title "Substantial effective sample sizes were required for external validation studies of predictive logistic regression models" @default.
- W2018195004 cites W165615793 @default.
- W2018195004 cites W1964619160 @default.
- W2018195004 cites W1980244389 @default.
- W2018195004 cites W1982989275 @default.
- W2018195004 cites W1989149883 @default.
- W2018195004 cites W1991179636 @default.
- W2018195004 cites W2002493616 @default.
- W2018195004 cites W2003614970 @default.
- W2018195004 cites W2008706768 @default.
- W2018195004 cites W2020794991 @default.
- W2018195004 cites W2021801488 @default.
- W2018195004 cites W2025028635 @default.
- W2018195004 cites W2046751517 @default.
- W2018195004 cites W2060050130 @default.
- W2018195004 cites W2060374620 @default.
- W2018195004 cites W2074703669 @default.
- W2018195004 cites W2084396669 @default.
- W2018195004 cites W2089463633 @default.
- W2018195004 cites W2094681746 @default.
- W2018195004 cites W2109325327 @default.
- W2018195004 cites W2113639857 @default.
- W2018195004 cites W2114869070 @default.
- W2018195004 cites W2123162799 @default.
- W2018195004 cites W2123912883 @default.
- W2018195004 cites W2124583370 @default.
- W2018195004 cites W2126436234 @default.
- W2018195004 cites W2129925362 @default.
- W2018195004 cites W2135542802 @default.
- W2018195004 cites W2140468511 @default.
- W2018195004 cites W2148371346 @default.
- W2018195004 cites W2157825442 @default.
- W2018195004 cites W2158190397 @default.
- W2018195004 cites W2163631231 @default.
- W2018195004 cites W2165139489 @default.
- W2018195004 cites W4237599469 @default.
- W2018195004 cites W4244216024 @default.
- W2018195004 doi "https://doi.org/10.1016/j.jclinepi.2004.06.017" @default.
- W2018195004 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15845334" @default.
- W2018195004 hasPublicationYear "2005" @default.
- W2018195004 type Work @default.
- W2018195004 sameAs 2018195004 @default.
- W2018195004 citedByCount "483" @default.
- W2018195004 countsByYear W20181950042012 @default.
- W2018195004 countsByYear W20181950042013 @default.
- W2018195004 countsByYear W20181950042014 @default.
- W2018195004 countsByYear W20181950042015 @default.
- W2018195004 countsByYear W20181950042016 @default.
- W2018195004 countsByYear W20181950042017 @default.
- W2018195004 countsByYear W20181950042018 @default.
- W2018195004 countsByYear W20181950042019 @default.
- W2018195004 countsByYear W20181950042020 @default.
- W2018195004 countsByYear W20181950042021 @default.
- W2018195004 countsByYear W20181950042022 @default.
- W2018195004 countsByYear W20181950042023 @default.
- W2018195004 crossrefType "journal-article" @default.
- W2018195004 hasAuthorship W2018195004A5009766171 @default.
- W2018195004 hasAuthorship W2018195004A5035355889 @default.
- W2018195004 hasAuthorship W2018195004A5056077573 @default.
- W2018195004 hasAuthorship W2018195004A5089919157 @default.
- W2018195004 hasConcept C105795698 @default.
- W2018195004 hasConcept C11413529 @default.
- W2018195004 hasConcept C114289077 @default.
- W2018195004 hasConcept C121332964 @default.
- W2018195004 hasConcept C129848803 @default.
- W2018195004 hasConcept C151956035 @default.
- W2018195004 hasConcept C152877465 @default.
- W2018195004 hasConcept C154945302 @default.
- W2018195004 hasConcept C155512373 @default.
- W2018195004 hasConcept C185592680 @default.
- W2018195004 hasConcept C19499675 @default.
- W2018195004 hasConcept C198531522 @default.
- W2018195004 hasConcept C2778755073 @default.
- W2018195004 hasConcept C33923547 @default.
- W2018195004 hasConcept C41008148 @default.
- W2018195004 hasConcept C43617362 @default.
- W2018195004 hasConcept C62520636 @default.
- W2018195004 hasConcept C83546350 @default.
- W2018195004 hasConcept C89128539 @default.
- W2018195004 hasConcept C96608239 @default.
- W2018195004 hasConcept C97931131 @default.
- W2018195004 hasConceptScore W2018195004C105795698 @default.
- W2018195004 hasConceptScore W2018195004C11413529 @default.
- W2018195004 hasConceptScore W2018195004C114289077 @default.
- W2018195004 hasConceptScore W2018195004C121332964 @default.
- W2018195004 hasConceptScore W2018195004C129848803 @default.
- W2018195004 hasConceptScore W2018195004C151956035 @default.
- W2018195004 hasConceptScore W2018195004C152877465 @default.
- W2018195004 hasConceptScore W2018195004C154945302 @default.
- W2018195004 hasConceptScore W2018195004C155512373 @default.