Lors d’une épreuve d’analyse sensorielle, des sujets préalablement entrainés dégustent desproduits et les notent à l’aide de différents descripteurs afin d’établir des profils descriptifs.De nombreuses méthodes statistiques (ACP, ANOVA…) permettent d’analyser les donnéesobtenues, afin notamment d’obtenir des résultats sur les performances des sujets, et dereprésenter les produits sur des cartes.Durant ce travail de thèse, ces méthodes statistiques ont été revisitées par l’exploitationd’un grand nombre de jeux de données issus de la SensoBase (base de données contenantplus de 1000 jeux de données de profil sensoriel). Différentes méthodes de cartographie etd’analyse des performances des sujets ont ainsi été comparées, puis des améliorations ontété proposées. La prise en compte du scaling (tendance d’un sujet à disperser ses notes plus–ou moins- que le panel) et l’adaptation de cartes à des modèles statistiques spécifiques ontété tout particulièrement étudiées.Ce travail a abouti à la documentation de certaines pratiques de l’analyse sensorielle et àl’établissement de recommandations à destination des animateurs de panel. Les différentesméthodes proposées dans cette thèse ont été implémentées sous R (librairies CVAPack,MAMCAP, MultiMAM), et sont donc directement utilisables par tout animateur de panel.Tous ces outils sont transposables dans d’autres domaines scientifiques utilisant l’êtrehumain comme instrument de mesure (psychologie, économie..). / During a sensory evaluation, trained subjects taste and score products on differentdescriptors in order to obtain a descriptive profile of these products. The data are thenanalyzed by several statistical methods (PCA, ANOVA…) in order to monitor the subjectperformances and map the product space.This work aims to revisit these statistical methods thanks to the analysis of a large number ofdatasets of sensory profiling data from the SensoBase (a database containing more than athousand of datasets). Different statistical methods of mapping and analysis ofperformances were compared, then improved. We mainly focused on the so-called scalingeffect (tendency of the subject to spread his scores more or less than the panel) and onmaps related to specific statistical models.This works led to some findings about sensory analysis and recommendations for panelleaders. Furthermore, different statistical tools (R packages such as CVAPack, MAMCAP andMultiMAM) were offered. These tools could also be useful in other scientific fields usinghuman beings as a measurement instrument (psychology, economics, marketing,… ).

Background: Laparoscopic surgery for colorectal cancer is a novel healthcare technology, for which much research evidence has been published. The objectives of this work were to compare the oncologic outcomes of this technology across different study types, and to define patterns of adoption on the basis of the literature. Methods: A comprehensive systematic review of the literature was conducted using 1) existing systematic reviews, 2) randomized controlled trials (RCTs), and 3) observational studies. Outcomes of interest were overall survival, and total lymph node harvest. Outcomes were compared for congruence. Adoption was evaluated by means of summary expert opinions in the literature. Results: 1) Existing systematic reviews were of low to moderate quality and displayed evidence of overlap and duplication. 2) Laparoscopy was not inferior to open surgery in terms of oncologic outcomes in any study type. 3) Oncologic outcomes from RCTs and observational studies were congruent. 4) Expert opinion in the literature has been supportive of this technology, paralleling the publication of large RCTs. Conclusions: The evaluation of laparoscopic surgery for colorectal cancer in RCTs and observational studies suggests that it is not inferior to open surgery. Adoption of this technology has paralleled RCT evidence.

The potential that telemedicine interventions may have in effectively delivering remote specialized cardiovascular care to large numbers of patients with heart diseases has recently come under question. In the first phase of this thesis, a systematic review and meta-analysis was conducted to compare the impact of a basic form of telemedicine that is regular patient follow-up by telephone, with usual care for individuals with coronary artery disease following their discharge. In the second phase of this thesis, a network meta-analysis, using Bayesian methods for multiple treatment comparisons, was conducted to compare the more complex forms of telemedicine for patients with heart failure. In the third and final phase of this thesis, a randomized controlled trial was designed to compare the impact of two forms of telemedicine, identified in the earlier two phases as being the most promising, on clinical outcomes, cardiac risk factors and patient reported outcomes.

Inflammatory bowel disease (IBD), which includes the subtypes Crohn's disease (CD) and ulcerative colitis (UC), is a common disease particularly in the Western world. IBD is characterised by inflammation of the small intestine and/or colon. The two subtypes affect different gut locations but both show an increased intestinal permeability or the 'leaky gut syndrome'. This led to the hypothesis that tight junction (TJ) proteins expressed in the epithelium may affect the intestinal permeability as a cause or effect of IBD.Initially, variants in the CARD15, IL23R and ATG16L1 genes, previously associated with an increased risk of IBD, were genotyped in a cohort of 500 IBD (295 CD and 205 UC) patients and 877 matched controls. These variants were significantly associated in our cohort. A random effects meta-analysis was undertaken on all previously reported CD associations with the variant rs2241880 from ATG16L1 (n=25, p=0.0017, OR: 1.36 95% CI 1.12-1.66) and with rs11209026 from IL23R (n=26, p=0.0006, OR: 0.37 95% CI 0.21-0.67), showing pooled odds ratios consistent with those reported in our cohort. Individuals carrying >1 CARD15 mutant variant were found to have a 2.5 fold increased risk of CD (p=0.0001). Candidate TJ proteins were chosen on the basis of previous reported associations and through the investigation of the claudin proteins which are abundant at TJs. Twenty one candidate genes were selected and 79 variants successfully genotyped in up to 1063 IBD (502 CD and 478 UC) and 870 control patients. Significant associations were detected with variants in the CLDN1, CLDN5 and CDH1 genes with CD; CLDN5, CLDN8 and CDH1 variants were associated to IBD; and the rs7791132 variant (between CLDN4 and ELN) and a CDH1 variant were associated to UC. The CLDN1 rs6809685 variant trended towards association in a Toronto ascertained IBD replication cohort (genotypic p=0.04, allelic p=0.06) suggesting this may be a novel IBD susceptibility variant. Small intestinal biopsies from CD patients with known rs6809685 genotypes showed a dose dependent reduced immunohistochemical staining of claudin 1 with carriage of the mutant G allele. Claudin 1 helps seal TJs and reduced levels may increase risk of CD.Peroxisome proliferator activator receptors (PPARs) can directly affect TJ proteins and could therefore affect intestinal permeability. Twelve PPARγ variants were genotyped in up to 1050 IBD (502 CD and 467 UC) and 725 control patients. Significant genotypic associations were found with the rs2067819 variant in CD (p=0.05) and IBD (p=0.02), and also the rs13099634 variant in UC (P=0.02). There was a strong gender difference particularly for rs2067819 and rs4135247, where allelic associations were highly significant and increased risk of IBD in men (p=0.01 and p=0.007 respectively). However no significant associations were found in the female cohort. Troglitazone a PPARα agonist increased Caco2 cell transepithelial electrical resistance (TEER), a marker of TJ integrity, and increased expression of claudins -3 and -4. In contrast, the PPARα antagonist GW6471 reduced the TEER without causing cell death and PPARγ ligands did not affect TEER measurements. In summary, using a robust cohort of cases and controls the data indicates that variants in genes encoding TJ proteins may affect susceptibility to IBD and that PPARs can regulate these proteins altering intestinal permeability.

Considerando a importância do setor elétrico para a economia de um país e os impactos ambientais causados por ele, o presente trabalho propõe a aplicação da Avaliação do Ciclo de Vida (ACV) para o cálculo do fator de emissão de gases de efeito estufa (GEE) da matriz de energia elétrica brasileira. Por meio da aplicação de uma revisão sistemática da literatura com meta-análise, foi possível identificar estudos que estimavam as emissões de GEE do ciclo de vida das tecnologias de geração da matriz brasileira e avaliar seus resultados conjuntamente. Os estudos selecionados apresentaram estimativas de emissões de GEE em um largo intervalo para uma mesma tecnologia. O procedimento estatístico adotado permitiu avaliar como as características das usinas podem afetar os resultados da ACV e reduzir a variabilidade das estimativas de emissões de GEE. Comparativamente, a energia eólica e a nuclear são as tecnologias que apresentam os menores fatores de emissão. Já entre os combustíveis fósseis, é a termeletricidade a gás natural que apresenta o melhor desempenho. As estimativas das emissões de hidrelétricas mostraram-se bastante variáveis, desde valores negativos até valores superiores aqueles encontrados para a termeletricidade a carvão mineral. Considerando que grande parte das emissões das hidrelétricas está relacionada com a decomposição da biomassa submersa em seus reservatórios, tem-se a indicação de que não há um consenso na metodologia utilizada para medi-las e que elas são extremamente dependentes das características dos reservatórios. Com isso, observa-se a importância de não negligenciar essas emissões na avaliação ambiental desses sistemas. O fator de emissão estimado para a matriz brasileira é de 125gCO2eq/kWh. Entende-se que no Brasil ainda exista uma série de desafios para que se construa uma base de dados de ACV com as características das tecnologias brasileiras. Portanto, com os resultados obtidos, espera-se contribuir para a inclusão da ACV nas análises ambientais de sistemas de geração de eletricidade, mostrando como essa abordagem é relevante para o planejamento energético que busque mitigar as emissões de GEE / Considering the electricity sector importance to a country economy and the environmental impacts that it causes, this work evaluates the adoption of Life Cycle Assessment (LCA) methodology to calculate the greenhouse gases (GHG) emission factor of the Brazilian electricity matrix. Using a literature systematic review together with a metaanalysis, this work identified studies that have estimated life cycle GHG emissions of electricity generation technologies of the Brazilian electricity matrix and combined the results of these studies. The selected studies provided estimates of GHG emissions in a wide range for the same technology. The statistical procedure adopted allowed to evaluate how characteristic factors of the generation technologies can affect the LCA results and to reduce the variability of GHG emissions estimates. In comparison, wind and nuclear power are the ones that have the lowest GHG emission factors. Among fossil fuels, natural gas-fired power plants have the best performance. Estimates of hydroelectric power plants emissions were quite variable, from negative values to values higher than those found for the coal-fired power plants. Whereas a large proportion of dams emissions is related to submerged biomass decomposition in the reservoirs, there is no consensus on the methodology used to measure such emissions which are dependent on reservoir characteristics. Still, one can notice the importance of not neglecting these issues in the environmental assessment of these systems. The emission factor estimated for the Brazilian electricity matrix is 125gCO2eq/kWh. It is understood that in Brazil still exists a number of challenges in order to build a LCA database with Brazilian technologies characteristics. Therefore, the results obtained in this work are expected to contribute to the inclusion of LCA in environmental analysis of electricity generation systems, showing how this approach is relevant to energy planning that seeks to mitigate GHG emissions

O principal objetivo deste trabalho é entender como coalizões de países em desenvolvimento são bem-sucedidas nas negociações do GATT (General Agreement on Tariffs and Trade) e da Organização Mundial de Comércio (OMC). Normalmente, sustenta-se que coalizões temáticas são mais bem-sucedidas do que grupos abrangentes (Higgot e Cooper, 1990; Cooper, Higgot e Nossal, 1990); e que coalizões com maiores recursos de poder são mais bem-sucedidas do que grupos fracos (Narlikar, 2003). Como hipótese alternativa, sugerimos que quanto maior for o grau de abertura comercial da coalizão - ((exportações+importações)/PIB) - maior será a chance de elas serem bemsucedidas nas negociações do GATT/OMC. Utilizaremos uma abordagem multimétodos (qualitativa e quantitativa) para realizar nossa pesquisa. Analisaremos um número (N) médio (entre 28 a 39 casos) de coalizões internacionais para descobrir quais as principais causas dos sucessos e fracassos destes grupos de países. Codificaremos a variável dependente (sucesso versus fracasso) conforme as descrições de casos particulares conduzidos por pesquisadores independentes. A variável independente categórica (coalizões temáticas versus abrangentes) será mensurada segundo procedimento semelhante. Por fim, o PIB agregado das coalizões (nossa medida de poder) e a abertura comercial agregada e abertura comercial média das coalizões (nossas medidas de abertura comercial) foram coletadas na base de dados Penn World Table 6.2. Os trabalhos que conhecemos sobre o tema utilizam a abordagem qualitativa de maneira pura, especialmente estudos de casso. Assim, até onde sabemos, este trabalho realizará a primeira análise estatística sobre o tema. / The main goal of this study is to understand how developing country coalitions obtain benefits in the GATT/WTO negotiations. Usually it is argued that issue-based coalitions are more successful than broad-based groups (Higgot and Cooper, 1990; Cooper, Higgot and Nossal, 1990), and that powerful coalitions are more successful than weaker ones (Narlikar, 2003). Alternatively, we suggest that the greater the degree of trade openness - ((exports+imports)/GDP) - the greater the chance that they will succeed in the GATT/WTO negotiations. We use a mix-method approach (qualitative and quantitative) to conduct our research. We will analyze a medium-N (from 28 to 39) cases of international coalitions to find out the main causes of coalitions\' successes and failures. We will code our dependent variable (success versus failure) according to cases studies conducted by independent researchers. Our categorical independent variable (issue-based versus broad-based coalitions) will be measured adopting the same procedure. Finally, aggregate GDP - our measure of power -, aggregate trade openness and mean aggregate trade openness of the coalitions - our measures of trade openness - were collected in the Penn World Table 6.2. To the best of our knowledge, this is the first statistical analysis on the subject.

A field covering generation and evaluation of the medical information is a broad one. During my doctoral studies I focused on medical information mainly using the evidence-based medicine (EBM) principles. This approach leads to the use of currently best available medical data in the treatment of patients. The EBM methodology might be classified into two basic groups: afferent and efferent one. The afferent part is focused on development of information resources whereas the efferent one promotes optimal use of these resources and a critical application of the retrieved facts. Due to practical reasons (to limit extent of the text) I concentrated in the thesis on only one field of my activity. The text deals mainly with afferent part of EBM, namely with formation of a qualitative and quantitative synthesis of knowledge through compilation of a systematic review and meta-analysis. Compilation of systematic reviews of prognostic markers is discussed in a detail. It is a field not appropriately covered in the Czech literature. The practical outcome documented in the thesis is a description of the systematic review of immunohistochemical prognostic markers in intracranial ependymomas that was created and published by our research team. Compilation of this systematic review and meta-analysis was one of the...

Les implications financières pour les acheteurs dans les fusions et acquisitions (F & A) ont été un sujet de fascination pour les chercheurs et les praticiens pendant des décennies. Malgré des recherches académique et commerciales approfondies visant à déterminer si et comment les acquéreurs obtiennent des résultats financiers à court et à long terme à la suite des fusions et acquisitions, la clarté de notre compréhension de ces questions demeure difficile à déterminer. Cette thèse de doctorat cherche à apporter plus de clarté à ces questions en examinant les interactions complexes entre plusieurs aspects clés des fusions et acquisitions. Le chapitre 1 examine comment l’expérience des acquéreurs influe sur le rendement à long terme au moyen de décisions clés avant et après la transaction et comment cette influence indirecte diffère dans les contextes nationaux et transfrontaliers. Le chapitre 2 explore les configurations des caractéristiques des transactions et des acquéreurs ainsi que les mécanismes de gouvernance d’entreprise des acquéreurs correspondant aux rendements anormaux cumulés des acquéreurs positifs (CAR). Le chapitre 3 étudie les effets interactifs entre les institutions formelles des pays d’accueil, les caractéristiques des acquéreurs et les mécanismes de gouvernance d’entreprise de l’acquéreur CAR. Enfin, le chapitre 4 examine l’influence des reportages d’affaires sur l’acquéreur CAR. / Financial implications for buyers in mergers and acquisitions (M&A) have been a topic of fascination with academics and practitioners for decades. Despite extensive business research dedicated toward investigating whether and how acquirers perform financially in the short and long terms following M&A, so far, the clarity of our understanding about these issues remains elusive. This doctoral thesis seeks to bring more clarity to these questions by examining complex interactions among several key aspects of M&A. Chapter 1 investigates how acquirer experience influences long-term performance through key pre- and post-transaction decisions and how such indirect influence differs in domestic and cross-border contexts. Chapter 2 explores the configurations of deal and acquirer characteristics as well as acquirer corporate governance mechanisms corresponding to positive acquirer cumulative abnormal returns (CAR). Chapter 3 investigates the interactive effects among host countries’ formal institutions, acquirer characteristics and corporate governance mechanisms on acquirer CAR. Finally, Chapter 4 examines the influence of business news reports on acquirer CAR.

Introducción: Las cicatrices post-acné son frecuentes y ocasionan un impacto negativo en la salud mental y en la calidad de vida de las personas. Recientemente el uso del plasma rico en plaquetas (PRP) se ha propuesto como parte de su manejo. Objetivos: Sintetizar la evidencia primaria con respecto a la eficacia y la seguridad del PRP como tratamiento adyuvante de las cicatrices atróficas post-acné. Métodos: Llevamos a cabo una revisión sistemática y metaanálisis de ensayos clínicos controlados. Realizamos búsqueda primaria en PUBMED-MEDLINE, WEB OF SCIENCE, WEB OF SCIENCE-MEDLINE, EMBASE, SCOPUS y CENTRAL; así como, en los registros de ensayos clínicos de EE.UU., Europa y de la Organización Mundial de la Salud (OMS) hasta diciembre del 2019. También buscamos en repositorios de literatura gris. Para la evaluación del riesgo de sesgo utilizamos el instrumento de la Colaboración Cochrane. Calculamos riesgos relativos (RR) mediante modelos de efectos aleatorios con un intervalo de confianza (IC) de 95%. Resultados: De un total de 547 registros, incluimos 14 estudios (n=429). Respecto a la mejoría clínica, evidenciamos superioridad del PRP como terapia adyuvante en los estudios que utilizaron escalas por cuartiles (RR=1,79; IC95% 1,39–2,31; I2=21,7%; p<0,001) y en los que emplearon la escala cualitativa de Goodman y Barón (RR=4,34; IC95% 1,60–11,76; I2=29,1%; p<0,001). Asimismo, en la evaluación de la satisfacción del paciente observamos superioridad en el grupo asignado a PRP (RR=1,51; IC95% 1,19-1,92; I2=51,6%; p=0,001). En el análisis por subgrupo, el grupo de PRP con microagujas obtuvo un mayor tamaño del efecto en cuanto a mejoría clínica y satisfacción del paciente (RR=2,66; IC95% 1,72–4,11; I2=0,0%; p<0,001; RR=1,67; IC95% 1,09–2,54; I2=69,9%; p<0,05, respectivamente). Además, el grupo con asignación “split face” obtuvo un mayor tamaño del efecto en cuanto a satisfacción del paciente (RR=1,58; IC95% 1,11–2,24; I2=65,6%; p<0,05). En cuanto a la seguridad, no realizamos metaanálisis por la alta heterogeneidad clínica y metodológica en el reporte de los efectos adversos. Siete publicaciones concluyeron que la duración e intensidad del eritema, edema o dolor fue menor en el grupo asignado a PRP, mientras que cuatro investigaciones informaron que no existió diferencia significativa para ambos grupos. Conclusiones: La evidencia proveniente de estudios clínicos controlados sugiere que el uso de PRP como terapia adyuvante es más eficaz y seguro que el recibir monoterapia en el manejo de las cicatrices atróficas post-acné. Asimismo, la combinación de PRP con microagujas presenta mayor efecto que con otras combinaciones. La aparición, duración e intensidad de efectos adversos es igual o menor con el uso del PRP como terapia adyuvante. Son necesarios nuevos ensayos clínicos aleatorizados controlados, con mayor tamaño de muestra, con protocolos estandarizados de uso del PRP y con escalas uniformes de medición de desenlaces. / Background: Acne scars are frequent and result in impairment of mental health and quality of life among affected individuals. The use of platelet-rich plasma (PRP) has been recently proposed as part of acne scarring management. Objective: To synthesize the primary evidence regarding the efficacy and safety of PRP as an adjuvant treatment of atrophic acne scars. Methods: We carried out a systematic review and meta-analysis of controlled clinical trials. We performed a primary search on PUBMED-MEDLINE, WEB OF SCIENCE, WEB OF SCIENCE-MEDLINE, EMBASE, SCOPUS, and CENTRAL. Moreover, we searched clinical trial registries from the United States of America, Europe, and of the World Health Organization (WHO) as well as gray literature repositories until December 2019. We used the Cochrane tool for assessing risk of bias. We estimated relative risks (RR) with 95% confidence interval (CI) using random-effects models. Results: From 547 registers, we included 14 clinical trials (n=429). The PRP group showed a significantly higher clinical improvement compared to the control group in the studies using quartile grading scales (RR=1.79; 95%CI 1.39–2.31; I2=21.7%; p<0.001) and in the studies using the qualitative scarring grading system of Goodman and Baron (RR=4.34; 95%CI 1.60-11.76; I2=29.1%; p<0.001). Furthermore, the patient satisfaction was higher in the PRP group compared to the control group (RR=1.52; 95%CI 1.20-1.93; I2=51.1%; p=0.001). In the subgroup analysis, the PRP group with microneedling obtained a larger effect size in terms of clinical improvement and patient satisfaction (RR=2.66; 95%CI 1.72-4.11; I2=0.0%; p<0.001; RR=1.67; 95%CI 1.09-2.54; I2=69.9%; p<0.05, respectively). In addition, the split-face group had a larger effect size in terms of patient satisfaction (RR=1.58, 95%CI 1.11-2.24; I2=65.6%; p<0.05). Regarding the safety of the PRP, we did not perform a meta-analysis due to the high clinical and methodological heterogeneity in the reporting of adverse effects. Seven publications concluded that the duration and intensity of erythema, edema, and pain was less in the PRP group, while four investigations reported no significant difference for both groups. Conclusions: Evidence from controlled clinical studies suggests that in the management of atrophic acne scars, the use of PRP as an adjuvant treatment is more effective and safer than receiving monotherapy. Also, the combination of PRP with micro-needles has more effect than other combinations. The occurrence, duration, and intensity of adverse effects are equal or less with the use of PRP as an adjuvant therapy compared to receiving monotherapy. There is a need for further randomized controlled trials with larger sample sizes, standardized protocols for the use of PRP, and uniform outcome measurement scales. / Tesis

PhD (Statistics) / Department of Statistics / There has been a growing interest in investment opportunities in Africa. The net foreign direct investment (FDI) to Sub-Saharan Africa has increased from $13 billion in 2004 to about $54 billion in 2015. Investing on the stock markets is one of such investment opportunities. Stock markets in Africa have realised growth in market capitalization, membership, value and volume traded due to an increase in investments. This level of growth in African stock markets has raised questions about their efficiency. This thesis examined the weak-form informational efficiency of African stock markets. The aim therefore of this thesis is to test the efficiency of African stock markets in the weak-form of the Efficient Market Hypothesis (EMH) for eight countries, namely, Botswana, Egypt, Kenya, Mauritius, Morocco, Nigeria, South Africa and Tunisia. Since, the researcher will be testing the weak-form of the EMH, the data to be used is on past price information on the markets of the eight countries. Data for the eight countries were obtained from DataStream for the period between August 28, 2000 to August 28, 2015. The data is for a period of 180 months which resulted in 3915 data points. Although there have been studies on the weak-form market efficiency of African stock markets, the efficiency conclusions on the markets have been mixed. This problem might be due to the methods used in the analyses. First, most of the methods used were linear in nature although the data generating process of stock market data is nonlinear and hence nonlinear methods maybe more appropriate in its analysis. Also these linear methods tested the efficiency of African markets in absolute form, however, an efficiency conclusion relying solely on absolute efficiency might be misleading because, stock markets become efficient with time due to improvements in the quality of information processing from reforms on the markets. The researcher solved this problem of using absolute frequency by comparing the results when the presence of long-memory in frequency and time domains of the markets were examined. The researcher used a semi-parametric estimator, the Local Whittle estimator to test for long-memory in frequency domain and the Detrended Fluctuation Analysis (DFA) to test for long-memory in time domain. The DFA method is suitable for both stationary and nonstationary time series which makes it to have more power over methods like the rescaled range analysis (R/S) in the estimation of Hurst exponent. Second, the researcher examined whether the markets were predictable under the Adaptive Market Hypothesis (AMH). The researcher employed the Generalised Spectral (GS) test to examine the Martingale difference hypothesis (MDH) of the markets. The Generalised spectral (GS) test is a non-parametric ii test designed to detect the presence of linear and nonlinear dependencies in a stationary time series. The GS test considers dependence at all lags. Third, because of the nonlinear nature in the data-generating process on the markets, the stationarity of the market returns under a nonlinear Exponential Smooth Threshold Autoregressive (ESTAR) model was examined. A nonlinear ADF unit root test against ESTAR and a modified Wald-type test against ESTAR in the analysis were employed. Fourth, the self-exciting threshold Autoregressive (SETAR) method was employed to model the returns when non-linear patterns were observed as a result of nonlinear data generating process on the markets. The literature on market efficiency of African stock markets has shown that variations exist in the study characteristics. There are variations in the method of analysis, type of test, type of data employed, time period chosen and the scope of analysis for the studies. The researcher therefore quantitatively reviewed previous studies by means of meta-analysis to identify which study characteristics affects efficiency conclusions of African markets using the mixed effects model. The findings showed the presence of long-memory in the returns of the stock markets when the whole sample was used. This made the markets weak-form inefficient, however, when the researcher tested for the persistence of long-memory through time, there were periods the markets were efficient in the weak-form. The memory effect was low in the South African market but high in the Mauritian market. Furthermore, it was observed that, the returns for Egypt, which were highly predictable when the whole data was analysed became not highly predictable when the rolling window approach of the GS test was used. Egypt had one of the lowest percentages of the windows that had a p-value less than 0.05 after South Africa. The results obtained from using the non-linear unit root tests on the logarithmic price series of the markets under study showed that, the markets were non-stationary and hence weak-form efficient under an ESTAR framework but for Botswana. Thus the markets were weak-form efficient when analysed using a non-linear method. This observation means that Africa’s foreign direct investment would have been increased over the years if the appropriate methods are used. This is because, over the years, studies on the weak-form efficiency African stock markets have ended with mixed conclusions with most of the markets being concluded to be weak-form inefficient as a result of the use of linear methods in the analysis. This finding, to us, has had an effect on investors commitments to Africa because the right methodology was not employed. iii The findings from modelling the returns under the non-linear SETAR model showed that, the SETAR model performs better than the standard AR(1) and AR(2) model for all the markets under study after the non-linear patterns were identified in the returns series. The SETAR (2,2,2) model is a threshold model, therefore, investors are able to move freely in search of higher opportunities between the low and high regimes. Investors main aim is to make profits, hence, the threshold model of SETAR gives them the freedom to move to a regime where the rate of returns is increasing unlike the standard AR(1) and AR(2) linear models where there are no switching of regimes. Finally, none of the study characteristics in the market efficiency studies was found to be significant in efficiency conclusions of African stock markets but the indicator for publication bias was significant. This means that there has been a change in attitude in recent years towards studies on informational market efficiency whose results do not support the Efficient Market Hypothesis (EMH), unlike the earlier years when the EMH was formulated and acclaimed to be one of the best propositions in economics. It was therefore concluded that when time-varying methods are used in analysing weak-form efficiency, the dynamics of the markets become known to investors for proper decision-making. Also, nonlinear methods should be used in order to reflect the nonlinear nature of data capturing on the stock markets / NRF