Model selection between the fixed-effects model and the random-effects model in meta-analysis

2020 ◽  
Vol 13 (4) ◽  
pp. 501-510
Author(s):  
Ke Yang ◽  
Hiu-Yee Kwan ◽  
Zhiling Yu ◽  
Tiejun Tong
Keyword(s):  
Model Selection ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Random Effects Model ◽  
Fixed Effects Model

The relationship between migraine and infant colic: A systematic review and meta-analysis

Cephalalgia ◽  
2014 ◽  
Vol 35 (1) ◽  
pp. 63-72 ◽  
Author(s):  
Amy A Gelfand ◽  
Peter J Goadsby ◽  
I Elaine Allen
Keyword(s):  
Systematic Review ◽  
Random Effects ◽  
Odds Ratio ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Primary Analysis ◽  
Infant Colic ◽  
The Relationship

Context Infant colic is a common and distressing disorder of early infancy. Its etiology is unknown, making treatment challenging. Several articles have suggested a link to migraine. Objective The objective of this article was to perform a systematic review and, if appropriate, a meta-analysis of the studies on the relationship between infant colic and migraine. Data sources Studies were identified by searching PubMed and ScienceDirect and by hand-searching references and conference proceedings. Study selection For the primary analysis, studies specifically designed to measure the association between colic and migraine were included. For the secondary analysis, studies that collected data on colic and migraine but were designed for another primary research question were also included. Data extraction Data were abstracted from the original studies, through communication with study authors, or both. Two authors independently abstracted data. Main outcomes and measures The main outcome measure was the association between infant colic and migraine using both a fixed-effects model and a more conservative random-effects model. Results Three studies were included in the primary analysis; the odds ratio for the association between migraine and infant colic was 6.5 (4.6–8.9, p < 0.001) for the fixed-effects model and 5.6 (3.3–9.5, p = 0.004) for the random-effects model. In a sensitivity analysis wherein the study with the largest effect size was removed, the odds ratio was 3.6 (95% CI 1.7–7.6, p = 0.001) for both the fixed-effects model and random-effects model. Conclusions In this meta-analysis, infant colic was associated with increased odds of migraine. If infant colic is a migrainous disorder, this would have important implications for treatment. The main limitation of this meta-analysis was the relatively small number of studies included.

scholarly journals A review and meta-analysis of red and processed meat consumption and breast cancer

2010 ◽  
Vol 23 (2) ◽  
pp. 349-365 ◽  
Author(s):  
Dominik D. Alexander ◽  
Libby M. Morimoto ◽  
Pamela J. Mink ◽  
Colleen A. Cushing
Keyword(s):  
Breast Cancer ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Random Effect ◽  
Red Meat ◽  
Processed Meat ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Meat Intake

The relationship between meat consumption and breast cancer has been the focus of several epidemiological investigations, yet there has been no clear scientific consensus as to whether red or processed meat intake increases the risk of breast cancer. We conducted a comprehensive meta-analysis incorporating data from several recently published prospective studies of red or processed meat intake and breast cancer. In the meta-analysis utilising data from the Pooling Project publication (includes data from eight cohorts) combined with data from nine studies published between 2004 and 2009 and one study published in 1996, the fixed-effect summary relative risk estimate (SRRE) for red meat intake (high v. low) and breast cancer was 1·02 (95 % CI 0·98, 1·07; P value for heterogeneity = 0·001) and the random-effects SRRE was 1·07 (95 % CI 0·98, 1·17). The SRRE for each 100 g increment of red meat was 1·04 (95 % CI 1·00, 1·07), based on a fixed-effects model, and 1·12 (95 % CI 1·03, 1·23) based on a random-effects model. No association was observed for each 100 g increment of red meat among premenopausal women (SRRE 1·01; 95 % CI 0·92, 1·11) but a statistically significant SRRE of 1·22 (95 % CI 1·04, 1·44) was observed among postmenopausal women using a random-effects model. However, the association for postmenopausal women was attenuated and non-significant when using a fixed-effects model (SRRE 1·03; 95 % CI 0·98, 1·08). The fixed- and random-effect SRRE for high (v. low) processed meat intake and breast cancer were 1·00 (95 % CI 0·98, 1·01; P value for heterogeneity = 0·005) and 1·08 (95 % CI 1·01, 1·16), respectively. The fixed- and random-effect SRRE for each 30 g increment of processed meat were 1·03 (95 % CI 1·00, 1·06) and 1·06 (95 % CI 0·99, 1·14), respectively. Overall, weak positive summary associations were observed across all meta-analysis models, with the majority being non-statistically significant. Heterogeneity was evident in most analyses, summary associations were sensitive to the choice of analytical model (fixed v. random effects), and publication bias appeared to have produced slightly elevated summary associations. On the basis of this quantitative assessment, red meat and processed meat intake does not appear to be independently associated with increasing the risk of breast cancer, although further investigations of potential effect modifiers, such as analyses by hormone receptor status, may provide valuable insight to potential patterns of associations.

scholarly journals Comparison of coated meshes for intraperitoneal placement in animal studies: a systematic review and meta-analysis

Hernia ◽  
2019 ◽  
Vol 24 (6) ◽  
pp. 1253-1261 ◽  
Author(s):  
H. Liu ◽  
S. van Steensel ◽  
M. Gielen ◽  
T. Vercoulen ◽  
J. Melenhorst ◽  
...  
Keyword(s):  
Systematic Review ◽  
Random Effects ◽  
Fixed Effects ◽  
Animal Studies ◽  
Meta Analysis ◽  
Adhesion Formation ◽  
Mean Difference ◽  
Random Effects Model ◽  
Fixed Effects Model

Abstract Purpose Laparoscopic intraperitoneal onlay mesh in hernia repair can result in adhesions leading to intestinal obstruction and fistulation. The aim of this systematic review is to compare the effects of mesh coatings reducing the tissue-to-mesh adhesion in animal studies. Methods Pubmed and Embase were systematically searched. Animal experiments comparing intraperitoneally placed meshes with coatings were eligible for inclusion. Only studies with comparable follow-up, measurements, and species were included for data pooling and subsequent meta-analysis. Results A total of 131 articles met inclusion criteria, with four studies integrated into one comparison and five studies integrated into another comparison. Compared to uncoated polypropylene (PP) mesh, PP mesh coated with hyaluronic acid/carboxymethyl cellulose (HA/CMC) showed significantly reduced adhesion formation at follow-up of 4 weeks measured with adhesion score of extent (random effects model, mean difference,−  0.96, 95% CI − 1.32 to − 0.61, P < 0.001, I2 = 23%; fixed effects model, mean difference,− 0.94, 95% CI − 1.25 to − 0.63, P < 0.001, I2 = 23%). Compared to PP mesh, polyester mesh coated with collagen (PC mesh) showed no significant difference at follow-up of 4 weeks regarding percentage of adhesion-area on a mesh, using random effects model (mean difference − 11.69, 95% CI − 44.14 to 20.76, P = 0.48, I2 = 92%). However, this result differed using fixed effects model (mean difference − 25.55, 95% CI − 33.70 to − 7.40, P < 0.001, I2 = 92%). Conclusion HA/CMC coating reduces adhesion formation to PP mesh effectively at a follow-up of 4 weeks, while the anti-adhesive properties of PC mesh are inclusive comparing all study data.

scholarly journals Comparison of efficacy of robotic surgery, laparoscopy, and laparotomy in the treatment of ovarian cancer: a meta-analysis

2019 ◽  
Vol 17 (1) ◽  
Author(s):  
Can Shi ◽  
Yingchun Gao ◽  
Yijun Yang ◽  
Lei Zhang ◽  
Juanpeng Yu ◽  
...  
Keyword(s):  
Ovarian Cancer ◽  
Robotic Surgery ◽  
Publication Bias ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Cochrane Library ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Estimated Blood Loss

Abstract Background We intended to compare the clinical effect of robotic surgery with laparoscopy and laparotomy in ovarian cancer treatment. Methods The included studies were retrieved from PubMed, Embase, and the Cochrane Library databases. The Methodological Index for Nonrandomized Studies (MINORS) was used to evaluate the study quality. Effect measures were presented with weighted mean difference (WMD)/odds ratio (OR) and 95% confidence interval (CI), and heterogeneity test was assessed using Q test and I2 statistics to determine the use of the random effects model or fixed effects model. Egger’s test was used to assess the publication bias. Results A total of eight studies was included in this meta-analysis with a MINORS score of 16–18. In the random effects model, estimated blood loss (EBL) of robotic surgery was significantly less compared with laparotomy (WMD = − 521.7027, 95% CI − 809.7816; − 233.6238). In the fixed effects model, length of hospital stay (LHS) (WMD = − 5.2225, 95% CI − 6.1485; − 4.2965) and postoperative complication (PC) (OR = 0.4710, 95% CI 0.2537; 0.8747) of robotic surgery were significantly less, and overall survival (OS) rate (OR = 6.4355, 95% CI 1.6722; 24.7678, P = 0.0070) of robotic surgery was significantly higher compared with laparotomy. There was no difference in the effect size of all variables between robotic surgery and laparoscopy. Meanwhile, a publication bias (t = 6.8290, P = 0.002405) was only identified for PC in robotic surgery and laparotomy groups; no publication bias was identified for the other variables. Conclusions Despite the above results, it failed to show oncological safety and recurrence by pathological stages or histologic types in this meta-analysis, and those confounding factors might affect the clinical outcome. Future meta-analyses with a larger number of eligible randomized controlled trial studies were needed to determine the most suitable treatment method for patients with different stages and types of ovarian cancer.

scholarly journals Association of Chronic Obstructive Pulmonary Disease With Arrhythmia Risks: A Systematic Review and Meta-Analysis

2021 ◽  
Vol 8 ◽  
Author(s):  
Xin Liu ◽  
Zhuohui Chen ◽  
Siyuan Li ◽  
Shuo Xu
Keyword(s):  
Chronic Obstructive Pulmonary Disease ◽  
Pulmonary Disease ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Current Data ◽  
Chronic Obstructive ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Obstructive Pulmonary Disease

Background: A large number of studies have shown that the arrhythmia risks may be the potential causes of death among chronic obstructive pulmonary disease (COPD) patients. However, the association of COPD with risks of arrhythmias has never been systematically reviewed. Therefore, we performed a meta-analysis to assess the relationship between COPD and arrhythmia risks.Methods: An updated systematic retrieval was carried out within the databases of Embase and PubMed until June 27, 2021.The random-effects model was used to pool studies due to the potential heterogeneity across the included studies. The risk ratios (RRs) with 95% confidence intervals (CIs) were regarded as effect estimates.Results: A total of 21 studies were included in our meta-analysis. In the pooled analysis by the random-effects model, the results showed that COPD was significantly related to the risk of atrial fibrillation (AF) (RR = 1.99, 95% CI: 1.46–2.70), ventricular arrhythmias (VA) (RR = 2.01, 95% CI: 1.42–2.85), and sudden cardiac death (SCD) (RR = 1.68, 95% CI: 1.28–2.21). The corresponding results were not changed after exclusion one study at a time. The pooled results were also stable when we re-performed the analysis using the fixed-effects model.Conclusions: Our current data suggested that COPD was associated with increased risks of AF, VA, and SCD.

scholarly journals Impact of Full IFRS, Accounting Standards for SMES and Company Demographics on Firms’ Return on Asset and Return on Equity using Panel Data Regression

2019 ◽  
Vol 8 (6S) ◽  
pp. 589-601 ◽  
Keyword(s):  
Least Squares ◽  
Random Effects ◽  
Fixed Effects ◽  
Random Effect ◽  
Ordinary Least Squares ◽  
Lagrangian Multiplier ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Real Property ◽  
The Impact

Due to globalization, markets are becoming more interconnected as the companies are engaged in doing cross-border offerings. Currently, competitions are intensified because Domestic organizations discover themselves competing with each nearby opposite numbers and worldwide companies. But one component that hinders SMEs is the need for reliable and similar monetary data. According to Abarca (2014), adoption of a high-quality and consistent set of accounting requirements is critical so as for the businesses to remain competitive in ASEAN member states. This paper ambitions to answer the query, what modified into the extent of the impact of compliance with full IFRS and IFRS for SMEs on profitability of agencies belong to real property enterprise? This paper moreover sought to decide whether there may be a sizeable distinction among the groups’ compliance with the overall PFRS and the PFRS for SMEs and to determine whether or now not there is a massive distinction among the companies’ financial normal overall performance earlier than and after the adoption of the PFRS for SMEs.Paired T-test have become employed in case you need to determine whether there is a big distinction between the agencies’ compliance with the entire PFRS and the PFRS for SMEs and to decide whether or not there may be a big difference some of the groups’ monetary performance earlier than and after the adoption of the PFRS for SMEs. Using STATA, the great appropriate version for every economic ratio on the subject of degree of compliance emerge as determined on. First, take a look at parm command became used to find out which most of the Least Squares Dummy Variable Regression Modes (LSDV1, LSDV2, LSDV3) underneath the Fixed Effects Model is the ideal version. Afterwards, Hausman Fixed Random Test changed into used to pick out out which is more suitable amongst Fixed Effects Model and Random Effects Model. If Fixed Effects Model modified into the more appropriate one, the Wald’s test turn out to be used to determine the best version among Fixed Effects Model and Ordinary Least Squares Model. On the alternative hand, if Random Effects Model became the more suitable one, the Breusch and Pagan Lagrangian Multiplier Test for Random Effect have become used to decide the satisfactory version amongst Random Effects Model and Ordinary Least Squares. Moreover, if Ordinary Least Squares became the splendid model, it is going to be in addition tested to check for heteroscedasticity and multicollinearity. White’s test became used to check for heterescedasticity and Variance Inflation Factor have become used to test if multicollinearity is gift. The results display that the adoption of PFRS for SMEs stepped forward the compliance of Philippine real property SMEs. However, no vast alternate became said inside the financial average performance of those companies (as measured with the resource of cross back on assets and go back on equity). This was further supported by the results of the panel regression. This means that despite having a relatively

scholarly journals FORECASTING MONEY DEMAND WITH THE INCLUSION OF FINANCIAL INNOVATION FOR THE ASSOCIATION OF SOUTHEAST ASIAN NATIONS: EVIDENCE FROM PANEL DATA ANALYSIS

2019 ◽  
Vol 1 (1) ◽  
pp. 59-74
Author(s):  
Payam Mohammad Aliha ◽  
Tamat Sarmidi ◽  
Fathin Faizah Said
Keyword(s):  
Time Series ◽  
Panel Data ◽  
Random Effects ◽  
Money Demand ◽  
Fixed Effects ◽  
Random Effects Model ◽  
Fixed Effects Model ◽  
Demand For Money ◽  
Financial Innovations ◽  
The Impact

This paper investigates the impact of financial innovations on the demand for money using a dynamic panel data for 10 ASEAN member states from 2004 to 2012 and attempt to forecast the demand for money during 2013 – 2016 to compare between forecasting performance of the fixed effects model with that of random effects model and also to compare the forecasting accuracy of dynamic forecasting and static forecasting obtained from these two models. An autoregressive model by definition is when a value from a time series is regressed on previous values from that same time series. There are two types of forecasting namely dynamic forecast and static forecast. “Dynamic forecast will take previously forecasted values while static forecast will take actual values to make next step forecast. Panel effects models assist in controlling for unobserved heterogeneity when this heterogeneity is constant over time and correlated (fixed effects) or uncorrelated (random effects) with independent variables. Hausman test indicates that the random-effects model is appropriate. We use the conventional money demand that is enriched with the number of automated teller machines (ATM) to proxy for the effect of financial innovations on money demand. By comparing the magnitude of “Root Mean Squared Error” (RMSE) as a benchmark for the two forecasts (0.1164 for dynamic forecast versus 0.0635 for static forecast) we simply find out that static forecast is superior to dynamic forecast meaning that static forecast provides more accurate forecast compared to a dynamic forecast for the fixed-effects model. Therefore, we conclude the static forecast on the basis of the random-effects model provides the most accurate forecasting. The estimation result of the chosen random-effects regression also indicates the estimated coefficient of ATM is not significant meaning that ATM does not impact money demand in ASEAN countries.

Meta Analysis of Study Habits and Academic Achievement

2021 ◽  
Vol 8 (4) ◽  
pp. 139-145
Author(s):  
M Elango ◽  
G Manimozhi
Keyword(s):  
College Students ◽  
Academic Achievement ◽  
Random Effects ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Study Habits ◽  
Random Effects Model ◽  
Inclusion Criteria ◽  
Analytical Review ◽  
The Relationship

A Meta analytical review investigated the study habits and academic achievement of school and college students. In this study the sample consists of twenty seven reviews based on study habits and academic achievement/ performance. The sample of the review based on inclusion criteria such as study habits and academic achievement. The review was collected by following details such as title of the study, Name of the author, Sample of the study, correlation Value and findings of the study. The Meta analysis is on the relationship between study habits and academic achievement. The main aim of the study is to find the fixed effects, random effects model values, Heterogeneity and forest plots. A Meta analysis of research paper was formed the conclusion.

MODELLING AND ESTIMATING INDIVIDUAL AND FIRM EFFECTS WITH COUNT PANEL DATA

2018 ◽  
Vol 48 (3) ◽  
pp. 1049-1078 ◽  
Author(s):  
Jean-François Angers ◽  
Denise Desjardins ◽  
Georges Dionne ◽  
François Guertin
Keyword(s):  
Panel Data ◽  
Random Effects ◽  
Fixed Effects ◽  
Parametric Model ◽  
Random Effects Model ◽  
Road Accidents ◽  
Fixed Effects Model ◽  
Insurance Pricing ◽  
Count Panel Data ◽  
Unobservable Factors

AbstractWe propose a new parametric model for the modelling and estimation of event distributions for individuals in different firms. The analysis uses panel data and takes into account individual and firm effects in a non-linear model. Non-observable factors are treated as random effects. In our application, the distribution of accidents is affected by observable and non-observable factors from vehicles, drivers and fleets of vehicles. Observable and unobservable factors are significant to explain road accidents, which mean that insurance pricing should take into account all these factors. A fixed effects model is also estimated to test the consistency of the random effects model.

scholarly journals The effect of cooperative learning on academic achievement: A meta-analysis on the relationship between the study group size and effect size İşbirlikli öğrenmenin başarıya etkisi: Çalışma grubu ile etki büyüklüğü arasındaki ilişkiye dair bir meta analiz çalışması

2017 ◽  
Vol 14 (1) ◽  
pp. 473 ◽  
Author(s):  
Murat Tuncer ◽  
Melih Dikmen
Keyword(s):  
Cooperative Learning ◽  
Random Effects ◽  
Effect Size ◽  
Fixed Effects ◽  
Meta Analysis ◽  
Experimental Studies ◽  
Study Group ◽  
Random Effects Model ◽  
Learning Groups ◽  
The Relationship

The purpose of this study was to determine the effect of cooperative learning on achievement and the relationship between the study group and effect size by means of meta-analysis. Meta-analysis is the method employed in order to statistically analyze the quantitative data collected in independent and multiple studies carried out on similar topics, and to reach a general judgment regarding the results of these studies.  Certain criteria were used in order to decide which researches would be included in the meta-analysis. Based on these criteria, it was decided to include 6 experimental studies in the meta-analysis. As a result, for the six studies, the effect size calculated within 95 % confidence interval has proved to be .518 based on random effects model. When two studies with the largest learning groups are excluded from the research respectively, firstly this effect size increases to .528, later it is calculated as .817 without any model discrimination. When effect size intervals in literature (≥ 0.5: strong, ≥ 0.3: moderate and  ≥ .01 weak) is taken into consideration, the effect of cooperative learning on achievement has been observed as  “Strong”. In addition to this, contrary to some views in literature, it has been seen that large learning groups have higher standard errors and a bigger effect size has been achieved when such studies have been excluded from the analysis. What is more, even the limitation of fixed effects model has disappeared, and the effect size calculated in fixed effects and random effects model has been balanced.  ÖzetBu araştırmanın amacı, işbirlikli öğrenmenin başarı üzerindeki etkisini ve çalışma grubu ile etki büyüklüğü arasındaki ilişkiyi meta-analiz yöntemiyle belirlemektir. Meta-analiz, benzer konularda yapılmış birbirinden bağımsız ve çok sayıda çalışmadan elde edilmiş sayısal verileri istatistiksel olarak analiz etme ve bu çalışmaların sonuçları hakkında genel bir yargıya varma yöntemidir. Hangi araştırmaların meta-analize dâhil edileceğine yönelik seçimlerin yapılmasında belirli ölçütler kullanılmıştır. Bu ölçütlere dayalı olarak 6 adet deneysel araştırmanın meta-analize dâhil edilmesine karar verilmiştir. Sonuç olarak; altı çalışma için %95 güven aralığında hesaplanan etki büyüklüğü rastgele etkiler modeline göre .518’dir. En büyük çalışma grubuna sahip iki araştırma sırasıyla araştırma dışında tutulduğunda bu etki büyüklüğü önce .528’ e çıkmakta, sonrasında ise model ayrımı olmaksızın .817 olarak hesaplanmaktadır. Alan yazındaki etki büyüklüğü aralıkları (≥ 0.5: güçlü, ≥ 0.3: orta düzey ve ≥ .01 zayıf) dikkate alındığında işbirlikli öğrenmenin başarı üzerindeki etkisi “Güçlü” bir etki olarak gözlenmiştir. Ayrıca alan yazındaki bazı görüşlerin aksine büyük çalışma gruplarının standart hatalarının daha yüksek olduğu, bu araştırmaların analiz dışında tutulması ile daha büyük etki büyüklüklerine ulaşıldığı, hatta sabit etki modelinin sınırlılığının bile ortadan kalktığı, sabit etki ve rastgele etki modelinde hesaplanan etki büyüklüğünün eşitlendiği görülmüştür.

Sign in / Sign up

Export Citation Format

Share Document