Abstracts in English, Xhosa and Southern Sotho / This study critically explores the link between internationalisation theories and outward foreign direct investment (FDI) – a linkage which is well documented in the literature. Numerous studies have established that the internationalisation process recognises both firm- and market-specific aspects, which greatly determine the direction of outward FDI in terms of volume and pattern. In this interaction, path dependency is determined by the intensity of overlapping aspects or linkages, from firm-level heterogeneity and host market aspects that direct investment patterns in terms of the latter, to the volumes of firm-level adjustments. Firm-level heterogeneity comprises those traits, which enable an individual firm to make an investment decision, select a market-entry strategy and create the competitive advantages that will sustain its investments. Macro-level or country-specific aspects encompass those traits or characteristics of host markets, which encourage FDI on the part of multinational enterprises (MNEs). Most studies overlook the path dependency of country- and firm-specific aspects, which are crucial to the internationalisation processes of international business, economics and trade. Academic studies either focus on macro- or micro-level aspects, without paying specific attention to the path dependency of expansion strategies. The present study attempts to fill these gaps in the existing body of knowledge, by investigating international business in these contexts. The rationale for undertaking this study was two-fold: first, FDI holds proven benefits for host markets, which include economic growth, industry spillover, human capital development and transitory tacit knowledge. From a firm-level perspective, outward FDI largely enhances the capacity of MNEs, prompting an increase in asset accumulation, market share and human capital development, the more efficient utilisation of resources and return on equity. In this study, an argument is presented for measuring the variables of both firm- and market-specific aspects, since most existing studies in this genre focus either on micro- or macro-level determinants, or totally overlook the importance of linkages. Second, no documented research has investigated the path dependency of expansion strategies, especially in Africa. Crucially, the importance of path dependency of South Africa’s outward investment has not been documented either. Further, existing evidence on the role the path dependency of expansion strategies plays in outward FDI are scarce, with even fewer studies following a sectorial approach. This study attempts to fill these academic research gaps by reflecting both firm- and market-level data from various sources for the period 1995–2015, using panel dynamic regression models. The study found that the linkages between firm heterogeneity (firm-level evidence) and market-level aspects create a path dependency of expansion strategies. MNEs adopt either joint ventures or wholly owned subsidiaries (or both) as market-entry strategies, but the decision is informed by the intensity of those firm heterogeneity aspects that allow them to exploit opportunities and mitigate risk in host markets. Notably, the intensity of path dependency seemingly varies from one industrial segment to the next. The impulse response approach delivered evidence that one standard deviation shock of firm-specific variables led to a moderate improvement in firm-level capacities in the short run, but a significant improvement in the long run. The same result was recorded for market-level aspects, with the intensity of the results varying from one industry to the next. The causality test attempted to explore the causal relationship between the study variables in both firm- and market-level aspects. Empirical evidence from the study indicates that the size of the firm and its capacity to utilise its resources efficiently, influence their investment in host markets. As regards market-specific aspects, the size of the economy, levels of industry and trade openness were found to have a causal effect on the inflow of FDI in host markets. The intensity of causal aspects was also found to vary from one industry to the next, due to variations in firm-level heterogeneity and their linkage in terms of aspects related to the host market. In sum, this study complements existing material on the subject of international business. / Olu phononongo luphicotha ikhonkco phakathi kweengcingane zamazwe ngamazwe kunye notyalo-mali ngokuthe ngqo oluphuma ngaphandle kumazwe asemzini (i-FDI) –indibaniselo ebhalwe kakuhle kwimiqulu yoncwadi. Izifundo ezininzi ezenziweyo ziye zaqinisekisa ukuba inkqubo yamazwe ngamazwe iyazamkela zombini inkampani- kunye nemiba ekhethekileyo yemarike, ezihlola kakhulukazi imikhombandlela (izikhokelo) ye-FDI yangaphandle ngokomthamo kunye nephatheni. Kule ntsebenziswano, indlela yokuxhomekeka ifunyanwa ngobungakanani bezinto ezisebenzelelanayo/ezingenanayo okanye izenzo zokuhlangana, ukusuka kwiintlobo-ntlobo zamanqanaba enkampani kunye neemfuno zabasingathe imicimbi yeendawo zokuthengisa (iimarike) iimpahla ezilawula iiphatheni zotyalo-imali ngokweyokugqibela, kwimilinganiselo yokulungelelaniswa kwenqanaba lwenkampani. Iintlobo-ntlobo zamanqanaba enkampani ziquka ezo mpawu, ezenza inkampani nganye yenze isigqibo sotyalo-mali, ikhethe isicwangciso sokungeniswa kwimarike kwaye siyile amathuba amahle okhuphiswano aya kugcina utyalo-mali. Inqanaba eliphezulu okanye iinkalo ezithile zelizwe zibandakanya ezo zimo okanye iimpawu zeemarike ezamkelekileyo, ezikhuthaza i-FDI kwiinkampani zamazwe ngamazwe (i-MNEs). Uninzi lwezifundo aziyiniki ngqalelo indlela yokuxhomekeka yelizwe kwimicimbi ekhethekileyo nebalulekileyo yenkampani kwiinkqubo zangokwamazwe oshishino lwamazwe ngamazwe, uqoqosho norhwebo. Uphando lwemfundo ephakamileyo lugxininisa kwiinkcukacha ezikwinqanaba eliphezulu okanye eliphantsi ngokunganiki ngqalelo kwindlela yokuxhomekeka yeendlela zokwandisa. Uphononongo lwangoku luzama ukuvala izikhewu/izikroba kulwazi olukhoyo., ngokuphanda ishishini lwamazwe ngamazwe kule meko. Ingqiqo ekwenzeni olu phando yahlulwe kubini: okokuqala, i-FDI inenzuzo eqinisekisiweyo kwabasingethe iimarike, ezibandakanya ukukhula koqoqosho, ukuchuma kwamashishini, ukuphuhliswa kwezakhono zabantu kunye nolwazi oludlulileyo lwezakhono. Ngakwicala lenqanaba lenkampani, i-FDI yangaphandle iphakamisa amandla e-MNE, ikhawulezisa ukunyusa uqokelelo lwempahla, isabelo semarike kunye nophuhliso lwabantu, ukusetyenziswa ngokufanelekileyo kwezixhobo kunye nokubuyela kubulungisa bokulingana. Kolu phononongo, impikiswano inikezelwe ukulinganisa iinguqu zombini yenkampani- kunye nemimiselo ethile yemarike, njengoko olunye uphando oluninzi olwenziweyo kolu hlobo lugxininisa koonobangela abakwizinga elisezantsi okanye eliphezulu, okanye kunganikwa ngqalelo tu kukubaluleka kwezenzo zokudibana / zokunxibelelana. Okwesibini, akukho phando lubhaliweyo oluphande indlela yokuxhomekeka kweendlela zokwandisa, ngakumbi e-Afrika. Ngokusesikweni, ukubaluleka kwendlela yokuxhomekeka yotyalo-mali lwangaphandle eMzantsi Afrika alukaze nalo lubhalwe phantsi. Ukongezelela, ubungqina obukhoyo kwindima yendlela yokuxhomekeka yeendlela zokwandisa kwi-FDI yangaphandle zinqabile, kwakunye nezifundo ezimbalwa ezilandela indlela yamacandelo. Olu phononongo luzama ukuzalisa izikroba zophando zemfundo ephakamileyo ngokuzibonakalisa zombini inkampani- kunye nedatha yamanqanaba emarike avela kwimithombo eyahlukeneyo yexesha lowe-1995-2015, usebenzisa iimodeli zepaneli ezinamandla zokubuy’umva. Uphononongo lufumanise ukuba ukudibana phakathi kweentlobo-ntlobo zenkampani (ubungqina bezinga lenkampani) kunye nemilinganiselo yezinga lemarike zidala indlela yoxhomekeko yeendlela zokukhula. Ii-MNE zamkela intsebenziswano ngokuhlangeneyo okanye bazibambele ngokwabo ngokupheleleyo (okanye zombini) njengeendlela zokungena kwimarike, kodwa isigqibo siphenjelelwa bubungakanani beentlobo-ntlobo zemicimbi yenkampani evumela ukuba baxhaphaze amathuba kwaye banciphise umngcipheko kwiimarike zenkampani. Ngokuphawulekayo, ubukhulu bokuxhomekeka wendlela yokuxhomekeka kukhangeleka kusahluka ukusuka kwicandelo elinye lozoshishino ukuya kwelinye elilandelayo. Indlela yokuphendula ngokungxama inikezele ubungqina bokuba ukuphazamiseka okusesikweni kwizinto eziguquguqukayo zenkampani ezikhethekileyo zikhokelele ekuphuculeni okusezingeni eliphakathi kwinqanaba kubungakanani benqanaba lenkampani ngexeshana, kodwa ukuphuculwa okubonakalayo nokubalulekileyo ekuhambeni kwexesha. Isiphumo esifanayo sabhalwa phantsi kwiinkalo zemarike, nobukhulu beziphumo zohluka ukusuka kwelinye ishishini ukuya kwelinye. Uvavanyo lwamaxesha athile luzame ukuphonononga ubudlelwane bamaxesha athile phakathi kwezifundo zezinto eziguquguqukayo kwiinkalo zombini inkampani –kunye nenqanaba lemarike/ neemeko zemarike. Ubungqina bamava obuvela kuphando lubonisa ukuba ubungakanani benkampani kunye namandla okusebenzisa uvimba wezixhobo ngokufanelekileyo, ziphembelela utyalo-mali kwiimarike zenkampani. Ngokubhekiselele kwimimandla ethile yemarike, ubungakanani boqoqosho, amazinga oshishino kunye nokuvuleka kwezorhwebo kufunyaniswe ukuba kunefuthe elenzekayo ngamaxesha athile ekungeneni kwe-FDI kubasingathi beemarike. Ubungakanani bemicimbi eyenzeka ngamaxesha athile yafunyanwa kwakhona ukuba yohlukile ukusuka kwelinye ishishini ukuya kwelinye, ngenxa yeenguqu kwiintlobo-ntlobo zeamanqanaba enkampani kunye nokudibana kwabo ngokwemiba enxulumene nabasingethe imarike. Kafutshane esi sifundo, sigcwalisa izixhobo ezikhoyo kwisihloko sezoshishino lamazwe ngamazwe. / Dinyakišišo tše di utolla ka tsinkelo kgokagano gareng ga diteori tša peyomaemong a boditšhabatšhaba le peeletšothwii ye e tšwago dinageng tša ka ntle (FDI) – e lego kgokagano yeo go ngwadilwego ka yona kudu ka dingwalweng. Dinyakišišo tše mmalwa di utollotše gore tshepedišo ya go bea maemong a boditšhabatšhaba e lemoga bobedi dilo tša difeme le tše di amanago le difeme, tšeo di laolago kudu fao FDI ya dinaga tša ka ntle e lebilego gona mabapi le bontši le mokgwa. Ka tirišanong ye, go tšea diphetho go ya ka maemo go laolwa ke bontši bja dilo tšeo di dirwago ka nako e tee goba dikamano, go tloga go go farologanya ditšweletšwa ka femeng le dilo tša mmaraka wa ka nageng tšeo di laolago mekgwa ya dipeeletšo mabapi le go ya ka mmaraka wa ka nageng, go ya go mehuta ye mentši ya dipeakanyo tša ka femeng. Go farologanya ditšweletšwa ka femeng go bopilwe ke diphetogo tše, tšeo di kgontšhago feme ye itšego go tšea sephetho sa mabapi le peeletšo, go kgetha maano a go tsena ka mmarakeng le go hlama menyetla ye mekaone yeo e tlago tšwetša pele peeletšo ya yona. Dikokwane tša ekonomi ye kgolo goba tša ka nageng di akaretša diphetogo tšeo goba dipharologantšhi tša mebaraka ya ka nageng, tšeo di hlohleletšago FDI ka karolong ya dikgwebo tša dinaga tša ka ntle (di-MNE). Dinyakišišo tše ntši di hlokomologile go tšea diphetho go ya ka maemo ga naga le ga dilo tša ka femeng ye itšego, tšeo di lego bohlokwa go tshepedišo ya peyomaemong a boditšhabatšhaba ya dikgwebo tša boditšhabatšhaba, diekonomi le kgwebišano. Dinyakišišo tša dirutegi di ka be di lebeletše kudu dilo tša ekonomi ye kgolo goba tša ye nnyane, ka ntle le go lebiša šedi ye kgolo go go tšea diphetho go ya ka maemo a boditšhabatšhaba ga maano a katološo. Dinyakišišo tše di leka go tlatša dikgoba tše ka tsebo ye e lego gona, ka go nyakišiša dikgwebo tša boditšhabatšhaba ka maemong a. Maikemišetšo a go dira dinyakišišo tše e bile a mabedi: sa mathomo, FDI e na le dikholego tšeo di tiišeditšwego go mebaraka ya ka dinageng, tšeo di akaretšago kgolo ya ekonomi, khuetšano ya diintasteri, tlhabollo ya bokgoni bja bašomi le phetišetšo ya tsebo ye e lego nyanyeng. Go ya ka maemong a difeme, FDI ye e tšwago dinageng tša ka ntle e godiša bokgoni bja di- MNE, ya hlohleletša koketšego ya khwetšo ya dithoto, ya kabelano ya mmaraka le ya tlhabollo ya bokgoni bja bašomi, tšhomišo ye kaone kudu ya methopo le go hwetša poelo go dikabelano. Ka mo dinyakišišong tše, go hlagišwa ntlha ya go ela diphapano tša bobedi dilo tša ka femeng le tša ka mmarakeng, ka ge bontši bja dinyakišišo tše di lego gona ka mo lekaleng le la dinyakišišo di lebeletše kudu tšeo di laolago ekonomi ye nnyane goba ye kgolo goba tšeo di hlokomologago ka botlalo bohlokwa bja dikgokagano. Sa bobedi, ga go dinyakišišo tšeo di ngwadilwego tšeo di nyakišišitšego go tšea diphetho go ya ka maemo ga maano a katološo, kudukudu ka Afrika. Se bohlokwa ke gore, bohlokwa bja go tšea diphetho go ya ka maemo ga peeletšo ya Afrika Borwa ya dinaga tša ka ntle ga se gwa ngwalwa le ge go le bjale. Godimo ga fao, bohlatse bjo bo lego gona ka ga mošomo wa go tšea diphetho go ya ka maemo fao go ralokago ka ga maano a katološo ka go FDI ya dinaga tša ka ntle e se bjo bontši, gomme go na le dinyakišišo tše mmalwa go latela mokgwa wo o lebeletšego makala. Dinyakišišo tše di leka go tlatša dikgoba tše tša dinyakišišo tša dirutegi ka go laetša tshedimošo ya bobedi ka maemong a difeme le ka mebarakeng go tšwa methopong ya mehutahuta go tloga ka mengwaga ya 1995–2015, ka go šomiša mekgwa ya kakanyo ya dikamano ye e fetogago. Dinyakišišo di hweditše gore dikamano gareng ga go farologanya ditšweletšwa (bohlatse bja ka maemong a difeme) le dilo tša maemo a ka mmarakeng di hlola go tšea diphetho go ya ka maemo ga maano a katološo. Di-MNE di šomiša masolo a mohlakanelwa goba dikhamphani tša ka fasana tšeo di laolwago ka botlalo (goba ka bobedi) bjalo ka maano a go tsena ka mmarakeng, eupša sephetho se laolwa ke bontši bja dilo tšeo tša go farologanya ditšweletšwa tšeo di di dumelelago go nyaka dibaka le go fokotša kotsi ka mebarakeng ya ka nageng. Seo se lemogilwego ke gore, bontši bja go tšea diphetho go ya ka maemo go bonala go fapane go ya ka karolo ya intasteri go ya go ye nngwe. Mokgwa wa go arabela kgoketšo wo o hlagišitšwego ka bohlatseng bja gore phapogo ya tlwaelo ya diphapano tša ka femeng e feleleditše ka kaonafalo ya magareng ya bokgoni bja difeme lebakeng le lekopana, eupša ka kaonafalo ye kgolo mo lebakeng le letelele. Dipoelo tše di swanago di begilwe ka go dilo tša maemo a ka mmarakeng, gomme bontši bja dipoelo tša fapana go ya ka diintasteri. Teko ya mathata yeo e bego e leka go utolla kamano ya tšeo di bakago se gareng ga phapano ya dinyakišišo ka go bobedi ka dilo tša ka femeng le tša ka mmarakeng. Bohlatse bja maitemogelo go tšwa ka mo dinyakišišong bo laetša gore bogolo bja feme le bokgoni bja yona bja go šomiša methopo ya yona gabotse ntle le mathata, di huetša peeletšo ya yona ka mebarakeng ya ka nageng. Mabapi le dilo tša ka mmarakeng, bogolo bja ekonomi, maemo a intasteri le go hloka sephiri ka kgwebišanong di bonwe di na le seabe sa go baka seemo go tseneng ga FDI ka mebarakeng ya ka nageng. Bontši bja dilo tše di bakago maemo go hweditšwe gape gore go fapana go ya ka diintasteri, ka lebaka la diphapano ka go farologanyo ya ditšweletšwa ka difemeng le kamano ya tšona mabapi le dilo tšeo di amanago le mmaraka wa ka nageng. Bjalo ka kakaretšo, dinyakišišo tše di tlaleletša dingwalwa tšeo di lego gona ka ga hlogotaba ya dikgwebo tša boditšhabatšhaba. / Business Management / D. Phil. (Business Management)

La randomisation mendélienne est une méthode d’instrumentation utilisant des instruments de nature génétique afin d’estimer, via par exemple la régression des moindres carrés en deux étapes, une relation de causalité entre un facteur d’exposition et une réponse lorsque celle-ci est confondue par une ou plusieurs variables de confusion non mesurées. La randomisation mendélienne est en mesure de gérer le biais de confusion à condition que les instruments utilisés soient valides, c’est-à-dire qu’ils respectent trois hypothèses clés. On peut généralement se convaincre que deux des trois hypothèses sont satisfaites alors qu’un phénomène génétique, la pléiotropie, peut parfois rendre la troisième hypothèse invalide. En présence d’invalidité, l’estimation de l’effet causal de l’exposition sur la réponse peut être sévèrement biaisée. Afin d’évaluer la potentielle présence d’invalidité lorsqu’un seul instrument est utilisé, Glymour et al. (2012) ont proposé une méthode qu’on dénomme ici l’approche de la différence simple qui utilise le signe de la différence entre l’estimateur des moindres carrés ordinaires de la réponse sur l’exposition et l’estimateur des moindres carrés en deux étapes calculé à partir de l’instrument pour juger de l’invalidité de l’instrument. Ce mémoire introduit trois méthodes qui s’inspirent de cette approche, mais qui sont applicables à la randomisation mendélienne à instruments multiples. D’abord, on introduit l’approche de la différence globale, une simple généralisation de l’approche de la différence simple au cas des instruments multiples qui a comme objectif de détecter si un ou plusieurs instruments utilisés sont invalides. Ensuite, on introduit les approches des différences individuelles et des différences groupées, deux méthodes qui généralisent les outils de détection de l’invalidité de l’approche de la différence simple afin d’identifier des instruments potentiellement problématiques et proposent une nouvelle estimation de l’effet causal de l’exposition sur la réponse. L’évaluation des méthodes passe par une étude théorique de l’impact de l’invalidité sur la convergence des estimateurs des moindres carrés ordinaires et des moindres carrés en deux étapes et une simulation qui compare la précision des estimateurs résultant des différentes méthodes et leur capacité à détecter l’invalidité des instruments. / Mendelian randomization is an instrumentation method that uses genetic instruments to estimate, via two-stage least squares regression for example, a causal relationship between an exposure and an outcome when the relationship is confounded by one or more unmeasured confounders. Mendelian randomization can handle confounding bias provided that the instruments are valid, i.e., that they meet three key assumptions. While two of the three assumptions can usually be satisfied, the third assumption is often invalidated by a genetic phenomenon called pleiotropy. In the presence of invalid instruments, the estimate of the causal effect of exposure on the outcome may be severely biased. To assess the potential presence of an invalid instrument in single-instrument studies, Glymour et al. (2012) proposed a method, hereinafter referred to as the simple difference approach, which uses the sign of the difference between the ordinary least squares estimator of the outcome on the exposure and the two-stage least squares estimator calculated using the instrument. Based on this approach, we introduce three methods applicable to Mendelian randomization with multiple instruments. The first method is the global difference approach and corresponds to a simple generalization of the simple difference approach to the case of multiple instruments that aims to detect whether one or more instruments are invalid. Next, we introduce the individual differences and the grouped differences approaches, two methods that generalize the simple difference approach to identify potentially invalid instruments and provide new estimates of the causal effect of the exposure on the outcome. The methods are evaluated using a theoretical investigation of the impact that invalid instruments have on the convergence of the ordinary least squares and two-stage least squares estimators as well as with a simulation study that compares the accuracy of the respective estimators and the ability of the corresponding methods to detect invalid instruments.

El presente trabajo estudia el impacto de la educación de la madre sobre la desnutrición crónica infantil para los años 2002 al 2016 en Perú. Este estudio se realizó mediante un análisis por diversos métodos del modelo econométrico de Variables Instrumentales. Se presentan los resultados para Mínimos Cuadrados en dos Etapas y Método Generalizado del Momento. También se realizan los modelos econométricos de Inclusión Residual en Dos Etapas y Probit para datos de panel. En la presente investigación se utilizó la Base de Datos Niños del Milenio. Se obtiene como resultado principal que la educación de la madre tiene un impacto negativo sobre la desnutrición crónica infantil por cada modelo econométrico para los años 2002 al 2016 en Perú. Palabras clave: Mínimos Cuadrados en dos Etapas; Método Generalizado del Momento; Modelo de Inclusión Residual en Dos Etapas; Probit Panel; Salud; Desnutrición Crónica Infantil; Riqueza; Logro Educativo; Perú; Niños del Milenio. / The document studies the impact of maternal education on stunted children from 2002 to 2016 in Peru. It is analyzed by two different Instrumental Variables Methods, Two-Stage Least Squares and General Method of Moments. Also, the document uses Two-Stage Residual Inclusion Model, and Dynamic Probit Model for panel data. The present investigation uses the Young Lives data base. The main result is that the mother's education has a negative impact on stunted children for each model from 2002 to 2016 in Peru. Keywords: Two-Stage Least Squares; General Method of Moments; Two-Stage Residual Inclusion; Panel Probit; Health; Wealth; Stunting; Education Attainment; Young Lives; Peru / Tesis

[pt] O objetivo desta dissertação é comparar através de um estudo de simulação diferentes estimadores de modelos não-lineares. Nós consideramos neste trabalho o estimador não-linear de mínimos quadrados em dois estágios (NL2SLS), o estimador não-linear de máxima verossimilhança de informação limitada (LIML) e o estimador com função controle (CF). Os resultados mostram que os estimadores CF e LIML possuem em geral uma performance superior ao do NL2SLS para os modelos selecionados. O trabalho considera uma aplicação de uma Curva de Phillips não-linear para a Economia Brasileira. / [en] The aim of this dissertation is to compare, in a simulation study, different nonlinear estimators for selected models. We consider the two-stage nonlinear least-squares (NL2SLS), the nonlinear limited information maximum likelihood (LIML), and the control function (CF) estimator. Our results show that usually either CF or LIML estimators perform better than the NL2SLS estimator for the selected models. In an application with real data, we consider the estimation a nonlinear Phillips Curve for Brazilian economy.

Submitted by Paulo Gustavo Grahl (pgrahl@fgvmail.br) on 2013-10-18T05:32:44Z No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Approved for entry into archive by ÁUREA CORRÊA DA FONSECA CORRÊA DA FONSECA (aurea.fonseca@fgv.br) on 2013-10-28T18:22:53Z (GMT) No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Approved for entry into archive by Marcia Bacha (marcia.bacha@fgv.br) on 2013-10-29T18:24:15Z (GMT) No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) / Made available in DSpace on 2013-10-29T18:25:35Z (GMT). No. of bitstreams: 1 DoutoradoPG_final.pdf: 23501670 bytes, checksum: 55b15051b9acc69ac74e639efe776fae (MD5) Previous issue date: 2012-12-22 / Esta dissertação concentra-se nos processos estocásticos espaciais definidos em um reticulado, os chamados modelos do tipo Cliff & Ord. Minha contribuição nesta tese consiste em utilizar aproximações de Edgeworth e saddlepoint para investigar as propriedades em amostras finitas do teste para detectar a presença de dependência espacial em modelos SAR (autoregressivo espacial), e propor uma nova classe de modelos econométricos espaciais na qual os parâmetros que afetam a estrutura da média são distintos dos parâmetros presentes na estrutura da variância do processo. Isto permite uma interpretação mais clara dos parâmetros do modelo, além de generalizar uma proposta de taxonomia feita por Anselin (2003). Eu proponho um estimador para os parâmetros do modelo e derivo a distribuição assintótica do estimador. O modelo sugerido na dissertação fornece uma interpretação interessante ao modelo SARAR, bastante comum na literatura. A investigação das propriedades em amostras finitas dos testes expande com relação a literatura permitindo que a matriz de vizinhança do processo espacial seja uma função não-linear do parâmetro de dependência espacial. A utilização de aproximações ao invés de simulações (mais comum na literatura), permite uma maneira fácil de comparar as propriedades dos testes com diferentes matrizes de vizinhança e corrigir o tamanho ao comparar a potência dos testes. Eu obtenho teste invariante ótimo que é também localmente uniformemente mais potente (LUMPI). Construo o envelope de potência para o teste LUMPI e mostro que ele é virtualmente UMP, pois a potência do teste está muito próxima ao envelope (considerando as estruturas espaciais definidas na dissertação). Eu sugiro um procedimento prático para construir um teste que tem boa potência em uma gama de situações onde talvez o teste LUMPI não tenha boas propriedades. Eu concluo que a potência do teste aumenta com o tamanho da amostra e com o parâmetro de dependência espacial (o que está de acordo com a literatura). Entretanto, disputo a visão consensual que a potência do teste diminui a medida que a matriz de vizinhança fica mais densa. Isto reflete um erro de medida comum na literatura, pois a distância estatística entre a hipótese nula e a alternativa varia muito com a estrutura da matriz. Fazendo a correção, concluo que a potência do teste aumenta com a distância da alternativa à nula, como esperado. / This dissertation focus on spatial stochastic process on a lattice (Cliff & Ord--type of models). My contribution consists of using Edgeworth and saddlepoint series to investigate small sample size and power properties of tests for detecting spatial dependence in spatial autoregressive (SAR) stochastic processes, and proposing a new class of spatial econometric models where the spatial dependence parameters that enter the mean structure are different from the ones in the covariance structure. This allows a clearer interpretation of models' parameters and generalizes the set of local and global models suggested by Anselin (2003) as an alternative to the traditional Cliff & Ord models. I propose an estimation procedure for the model's parameters and derive the asymptotic distribution of the parameters' estimators. The suggested model provides some insights on the structure of the commonly used mixed regressive, spatial autoregressive model with spatial autoregressive disturbances (SARAR). The study of the small sample properties of tests to detect spatial dependence expands on the existing literature by allowing the neighborhood structure to be a nonlinear function of the spatial dependence parameter. The use of series approximations instead of the often used Monte Carlo simulation allows a simple way to compare test properties across different neighborhood structures and to correct for size when comparing power. I obtain the power envelope for testing the presence of spatial dependence in the SAR process using the optimal invariant test statistic, which is also locally uniformly most powerful invariant (LUMPI). I have found that the LUMPI test is virtually UMP since its power is very close to the power envelope. I suggest a practical procedure to build a test that, while not UMP, retain good power properties in a wider range for the spatial parameter when compared to the LUMPI test. I find that power increases with sample size and with the spatial dependence parameter -- which agrees with the literature. However, I call into question the consensus view that power decreases as the spatial weight matrix becomes more densely connected. This finding in the literature reflects an error of measure because the hypothesis being compared are at very different statistical distance from the null. After adjusting for this, the power is larger for alternative hypothesis further away from the null -- as one would expect.