Two-tailed asymptotic inferences for the odds ratio in cross-sectional studies : evaluation of fifteen old and new methods of inference
Various asymptotic methods of obtaining a confidence interval (CI) for the odds ratio (OR) have been proposed. Surprisingly, insofar as we know, the behavior of these methods has not been evaluated for data proceeding from a cross-sectional study (multinomial sampling), but only for data that originate in a prospective or retrospective study (two independent binomials sampling). The paper evaluates 15 different methods (10 classic ones and 5 new ones). Because the CI is obtained by inversion in θ of the two-tailed test for H0(θ): OR =[Formula: see text] (null hypothesis), this paper evaluates the tests for various values of θ, more than the CIs that are obtained. The following statements are valid only for the two-tailed inferences based on 20 ≤ n ≤ 200 and 0.05≤ OR≤20, since these are the limitations of the study. The two best methods are the classic Cornfield chi-squared method for 0.2≤ OR≤5 and, in other cases, the new method of Sterne for chi-squared; but the adjusted likelihood ratio method is a good alternative to the two previous methods, especially to the first when the sample size is large. The three methods require iterative calculations to obtain the CI. If one is looking for methods that are simple to apply (that is, ones that admit a simple, explicit solution), the best option is the Gart logit method for 1/3≤ OR≤3 and, if in other cases, the Agresti logit method. The Cornfield chi-squared and Gart logit methods should not be used outside the specified interval OR. The paper also selects the best methods for realizing the classic independence test (θ = 1).
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2020 |
---|---|
Erschienen: |
2020 |
Enthalten in: |
Zur Gesamtaufnahme - volume:30 |
---|---|
Enthalten in: |
Journal of biopharmaceutical statistics - 30(2020), 5 vom: 02. Sept., Seite 900-915 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Andrés, Antonio Martín [VerfasserIn] |
---|
Links: |
---|
Anmerkungen: |
Date Completed 02.08.2021 Date Revised 02.08.2021 published: Print-Electronic Citation Status MEDLINE |
---|
doi: |
10.1080/10543406.2020.1757691 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM310022150 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM310022150 | ||
003 | DE-627 | ||
005 | 20231225135206.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2020 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1080/10543406.2020.1757691 |2 doi | |
028 | 5 | 2 | |a pubmed24n1033.xml |
035 | |a (DE-627)NLM310022150 | ||
035 | |a (NLM)32419581 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Andrés, Antonio Martín |e verfasserin |4 aut | |
245 | 1 | 0 | |a Two-tailed asymptotic inferences for the odds ratio in cross-sectional studies |b evaluation of fifteen old and new methods of inference |
264 | 1 | |c 2020 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 02.08.2021 | ||
500 | |a Date Revised 02.08.2021 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status MEDLINE | ||
520 | |a Various asymptotic methods of obtaining a confidence interval (CI) for the odds ratio (OR) have been proposed. Surprisingly, insofar as we know, the behavior of these methods has not been evaluated for data proceeding from a cross-sectional study (multinomial sampling), but only for data that originate in a prospective or retrospective study (two independent binomials sampling). The paper evaluates 15 different methods (10 classic ones and 5 new ones). Because the CI is obtained by inversion in θ of the two-tailed test for H0(θ): OR =[Formula: see text] (null hypothesis), this paper evaluates the tests for various values of θ, more than the CIs that are obtained. The following statements are valid only for the two-tailed inferences based on 20 ≤ n ≤ 200 and 0.05≤ OR≤20, since these are the limitations of the study. The two best methods are the classic Cornfield chi-squared method for 0.2≤ OR≤5 and, in other cases, the new method of Sterne for chi-squared; but the adjusted likelihood ratio method is a good alternative to the two previous methods, especially to the first when the sample size is large. The three methods require iterative calculations to obtain the CI. If one is looking for methods that are simple to apply (that is, ones that admit a simple, explicit solution), the best option is the Gart logit method for 1/3≤ OR≤3 and, if in other cases, the Agresti logit method. The Cornfield chi-squared and Gart logit methods should not be used outside the specified interval OR. The paper also selects the best methods for realizing the classic independence test (θ = 1) | ||
650 | 4 | |a Comparative Study | |
650 | 4 | |a Journal Article | |
650 | 4 | |a Research Support, Non-U.S. Gov't | |
650 | 4 | |a Asymptotic confidence intervals | |
650 | 4 | |a Fieller method | |
650 | 4 | |a Sterne method | |
650 | 4 | |a chi-squared method | |
650 | 4 | |a likelihood ratio method | |
650 | 4 | |a logit and inverse sine transformations | |
650 | 4 | |a odds ratio | |
700 | 1 | |a García, Juan Miguel Tapia |e verfasserin |4 aut | |
700 | 1 | |a Moreno, Francisco Gayá |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Journal of biopharmaceutical statistics |d 1991 |g 30(2020), 5 vom: 02. Sept., Seite 900-915 |w (DE-627)NLM012811432 |x 1520-5711 |7 nnns |
773 | 1 | 8 | |g volume:30 |g year:2020 |g number:5 |g day:02 |g month:09 |g pages:900-915 |
856 | 4 | 0 | |u http://dx.doi.org/10.1080/10543406.2020.1757691 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 30 |j 2020 |e 5 |b 02 |c 09 |h 900-915 |