Bakgrund: Det finns olika investeringsstrategier som investerare kan använda sig av, att investera i låga multiplar är en strategi som har studerats väl. Genom att använda sig av låga multiplar kan investerare finna undervärderade bolag som på sikt genererar en överavkastning gentemot marknaden.  Syfte: Studiens syfte är att analysera hur väl P/E, P/B, P/S, EV/EBIT, EV/EBITDA och EV/S multiplarna skulle kunna appliceras som investeringsstrategi på Stockholmsbörsen. Vidare ämnar studien åt att analysera om det är möjligt att generera en högre avkastning än vad indexet OMXSPI har avkastat under tidsperioden 2008-2018. Metod: Studien använder sig av en kvantitativ forskningsstrategi där två portföljer för respektive multipel har sammanställts. Portföljerna viktas om årligen och både den verkliga och den ackumulerade avkastningen beräknas fram. Vidare utvärderas portföljerna enligt utvärderingsmåtten Sharpekvot, M^2, Treynorkvot och Jensens Alpha. Resultat: Investeringsstrategin är implementerbar för tre av sex multiplar. Låga P/B, EV/EBIT och EV/EBITDA genererade en överavkastning och slog både index samt respektive hög portfölj. Medan för de resterande multiplarna P/E, P/S och EV/S resulterade det i att investeringsstrategin inte är implementerbar. EV/S hade den högsta riskjusterade överavkastning och presterade bäst av samtliga sex multiplar. Studieresultatet för samtliga multiplar kan statistiskt säkerställas med en signifikansnivå på 5%. Den månatliga portföljavkastningen är inte slumpmässig, utan marknadsavkastningen har en viss påverkan. / Background: There are several investment strategies investors can use, where the strategy to invest in low multiples is well studied. By using low multiples investors can find undervalued companies to generate an excess return. Previous studies have been focusing on the P/E and EV/EBITDA- multiples, and not as much on other used multiples in relative valuation. Therefore an interest exists to also analyze multiples such as P/B, P/S, EV/EBIT and EV/S. Purpose: The study’s purpose is to analyze how well the multiples P/E, P/B, P/S, EV/EBIT, EV/EBITDA and EV/S can be applied as an investment strategy in the Stockholm Stock Exchange. Furthermore the study aim to analyze the possibility to generate a higher return than the index OMXSPI during the time period 2008-2018. Method: The study uses a quantitative research strategy, where two portfolios for each multiple has been created. The portfolio has been reinvested once a year, both the real and accumulated return was calculated. Also, the portfolios’ performance has been evaluated by adjusting it to risk by using the Sharpe ratio, M^2 , Treynor ratio and Jensen’s Alpha. Result: The investment strategy can be implemented for three of six multiples. The low P/B, EV/EBIT and EV/EBITDA generated a higher return than both index and their respective high portfolio. The other multiples P/E, P/S and EV/S cannot be used as an investment strategy. The high EV/S portfolio had the highest risk adjusted excess return meanwhile P/S had the highest accumulated return. The result of all multiples has been found to be statistically significant, therefore the market return has an effect on the portfolios’ monthly return.

Este trabalho de pesquisa teve por objetivo avaliar o efeito do descarte de lodo aeróbio no desempenho do reator UASB. Em paralelo, foi verificado o efeito dessa operação na qualidade do efluente final da ETE, no desempenho operacional da unidade de pós-tratamento e nas características do biogás produzido no reator UASB. De modo geral, não houve impacto negativo no processo de digestão anaeróbia em termos de DQO filtrada no efluente, que foi semelhante quando o afluente era composto essencialmente de esgoto bruto, na ordem de 60 a 80 mg/L. Por outro lado, os resultados indicaram uma tendência de redução da qualidade do efluente do reator UASB em termos de DQO total e sólidos em suspensão, quando da alimentação desse reator com excesso de lodo aeróbio. Em 2016, a concentração média de DQO total no efluente foi da ordem de 529 mg/L e a concentração média de SST foi na ordem de 672 mg/L quando o afluente do reator UASB era composto com excesso de lodo aeróbio, ao passo que, sem excesso de lodo aeróbio no afluente, a DQO total média no efluente foi de 269 mg/L e a concentração média de SST foi de 349 mg/L. Porém, essa redução da qualidade pareceu estar relacionada a uma sobreposição de efeitos associado a presença de lodo aeróbio no afluente, e a sistemática adotada na estação para remoção de lodo anaeróbio do reator UASB. Quando não havia remoção de excesso de lodo anaeróbio de fundo do reator UASB, a qualidade do efluente se manteve durante a maior parte do tempo monitorado compatível com os períodos onde o afluente do reator UASB era composto essencialmente de esgoto bruto, indicando que o lodo de fundo, mais concentrado, exerce papel fundamental na retenção do lodo aeróbio na zona de digestão do reator UASB. Essa constatação se mostrou mais explícita quando as operações de descarte de excesso de lodo aeróbio e remoção de lodo anaeróbio do reator UASB ocorriam em períodos distintos. Nessa condição, a concentração média de DQO total no efluente foi de 254 mg/L e a concentração média de SST foi de 314 mg/L, ao passo que sem lodo aeróbio no afluente do reator UASB, a concentração média de DQO total no efluente foi de 191 mg/L e a concentração de SST foi de 258 mg/L. As características do biogás produzido variaram quando o sistema era alimentado com excesso de lodo aeróbio no afluente, apresentando, nessa condição, maior concentração de CO2 e menor concentração de H2S. Não houve, alteração da concentração média de metano quando da presença de excesso de lodo aeróbio no afluente. Outro aspecto observado foi a falta de correlação entre os parâmetros DQO removida e produção volumétrica de biogás quando o afluente do reator UASB continha excesso de lodo aeróbio. Na ausência de excesso de lodo aeróbio no afluente, houve correlação entre essas variáveis, porém uma correlação fraca, na ordem de 0,40, que pode ser explicada por uma captação ineficaz do biogás produzido, devido a vazamentos pela linha de quebra-escuma do separador trifásico, e por perdas de metano dissolvido com o efluente. Com relação ao impacto na unidade de pós-tratamento, a baixa capacidade de remoção de lodo de excesso da ETE devido a dificuldades de ordem operacional e contratual, condicionaram a operação do sistema de lodo ativado com elevadas concentrações de sólidos em suspensão. Nessa condição, o sistema de aeração apresentou baixa eficiência de transferência de oxigênio para o licor misto, o que impactou negativamente no consumo de energia elétrica do processo de lodo ativado, aumentando o custo operacional da ETE. A qualidade do efluente final da ETE, de modo geral, não foi impactada pela operação de descarte de lodo aeróbio no reator UASB, tendo mantido durante a maior parte do tempo DQO média na ordem de 40 mg/L no efluente e concentração de sólidos em suspensão totais em média na ordem de 25 mg/L. A nitrificação e desnitrificação foi variável ao longo do período do estudo, sendo que o processo apresentou eficiência média de remoção de nitrogênio em torno de 66%. / The objective of this research was to evaluate the effect of aerobic sludge discard on UASB reactor performance. In parallel, the effect of this operation on the final effluent quality of the WWTP, the operational performance of the post-treatment plant and the characteristics of the biogas produced in the UASB reactor were verified. In general, there was no negative impact on the anaerobic digestion process in terms of dissolved COD in the effluent, which was similar when the influent was composed essentially of raw sewage, in the order of 60 to 80 mg / L. On the other hand, the results indicated a tendency to reduce the quality of the UASB reactor effluent in terms of total COD and suspended solids when feeding this reactor also with excess aerobic sludge. In 2016, the mean concentration COD in the effluent was of the order of 529 mg / L and the mean concentration SST was 672 mg / L when the UASB reactor inflow was composed of excess aerobic sludge, while, without excess of aerobic sludge in the influent, the mean COD in the effluent was 269 mg / L and the mean concentration of SST was 349 mg / L. However, this quality reduction seemed to be related to an overlap of effects associated with the presence of aerobic sludge in the influent, and the system adopted at the WWTP to remove anaerobic sludge from the UASB reactor. When there was no removal of excess anaerobic bottom sludge from the UASB reactor, the quality of the effluent remained for most of the monitored time consistent with the periods where the UASB reactor inflow was essentially composed of raw sewage, indicating that the sludge from deep, more concentrated, plays a fundamental role in the retention of aerobic sludge in the digestion zone of the UASB reactor. This finding was more explicit when the operations of excess aerobic sludge and anaerobic sludge removal from the UASB reactor occurred at different periods. In this condition, the mean COD concentration in the effluent was 254 mg / L and the mean SST concentration was 314 mg / L, whereas without aerobic sludge in the UASB reactor, the mean total COD concentration in the effluent was 191 mg / L and the concentration of SST was 258 mg / L. The characteristics of the biogas produced varied when the system was fed with excess of aerobic sludge in the influent presenting, in this condition, a higher concentration of CO2 and a lower concentration of H2S. There was, fortunately, no change in the average concentration of methane when there was an excess of aerobic sludge in the influent. Another important aspect observed was the lack of correlation between the COD removal parameters and volumetric biogas production when the UASB reactor inflow contained excess aerobic sludge. In the absence of excess aerobic sludge in the influent, there was a correlation between these variables, but a weak correlation, in the order of 0.40, which can be explained by an inefficient capture of the biogas produced, due to leaks by the three-phase separator, and for losses of methane dissolved with the effluent. Regarding the impact on the post-treatment unit, the low sludge removal capacity of the WWTP due to operational and contractual difficulties, conditioned the operation of the activated sludge system with high concentrations of suspended solids. In this condition, the aeration system presented low oxygen transfer efficiency for the mixed liquor, which negatively impacted the electric power consumption of the activated sludge process, increasing the operational cost of the WWTP. The final effluent quality, in general, was not affected by the aerobic sludge discharge operation in the UASB reactor, having maintained for the most part the average COD in the order of 40 mg / L in the effluent and the solids concentration in suspension on average in the order of 25 mg / L. Nitrification and denitrification were variable throughout the study period, with the average nitrogen removal efficiency around 66%.

Bakgrund & Problemformulering: Momentumeffekten på fondmarknaden är ett relativt outforskat område där dess existens på senare tid har blivit omtvistad. Eftersom kunskapen om pensionssparande och det svenska pensionssystemet är låg, samtidigt som de sociala skyddsnäten i samhället minskar är det viktigt att undersöka om momentumstrategier kan ge överavkastning för privatpersoners pensionssparande. Således ställs frågan: Kan momentumstrategier skapa överavkastning på fondmarknaden? Syfte: Syftet med uppsatsen är att undersöka momentumeffekten på fondmarknaden och om momentumstrategier kan utnyttjas av svenska pensionssparare för att skapa överavkastning i PPM-systemet. Metod: Uppsatsen har ett kvantitativt tillvägagångssätt och en deduktiv utgångspunkt tillämpas. För att undersöka momentumeffekten på fondmarknaden tillämpas en multipel regressionsanalys med Fama French-Trefaktormodell, samt Sharpekvot. Uppsatsens urval är PPM-fonder under perioden 2010-2017. Slutsatser: Uppsatsen finner inget statistiskt stöd för en momentumeffekt på fondmarknaden genom Fama French-Trefaktormodell. Detta är ett tecken på att fondmarknaden kan vara svagt effektiv då historisk information inte har kunnat användas för att skapa riskjusterad överavkastning. Uppsatsen finner således ingen momentumeffekt för fondmarknaden efter finanskrisen 2008, trots att en momentumeffekt har kunnat påvisas dessförinnan inom tidigare forskning. Med hänsyn till det har författarna anledning att misstänka att marknadens effektivitet kan variera, vilket skulle kunna förklaras av den Adaptiva Marknadshypotesen. / Background & Problem: The momentum effect in the fund market is relatively unexplored were its existence has been controversial. Due to the lack of knowledge in retirement savings and the Swedish Premium Pension Agency system, alongside the weakening of a social safety net, it is important to examine if momentum strategies give excess returns and can be used for retirement savings. Therefore, the authors question: Do momentum strategies give excess returns in the fund market? Purpose: The aim of the thesis is to examine the momentum effect in the fund market and if momentum strategies can be used to create excess return in the Premium Pension Agency system. Method: The thesis takes a deductive research approach with a quantitative methodology. To examine the momentum effect in the fund market, a multiple regression analysis model from Fama French-Three factor model is applied, and the Sharpe ratio. The sample for the study is Swedish Premium Pension Agency funds, which is examined over the period of 2010-2017. Conclusions: The thesis does not find support for a momentum effect in the fund market through the Fama French-Three factor model. This indicates that the fund market is weak form efficient, as historical information cannot be used to create risk adjusted excess return. Thus, the thesis does not find a momentum effect for the fund market after the financial crisis in 2008, even though a momentum effect is proven to exist before then. In view of this, the authors have reason to suspect the market efficiency to vary, which could be explained by the Adaptive Market Hypothesis.

Este trabalho de pesquisa teve por objetivo avaliar o efeito do descarte de lodo aeróbio no desempenho do reator UASB. Em paralelo, foi verificado o efeito dessa operação na qualidade do efluente final da ETE, no desempenho operacional da unidade de pós-tratamento e nas características do biogás produzido no reator UASB. De modo geral, não houve impacto negativo no processo de digestão anaeróbia em termos de DQO filtrada no efluente, que foi semelhante quando o afluente era composto essencialmente de esgoto bruto, na ordem de 60 a 80 mg/L. Por outro lado, os resultados indicaram uma tendência de redução da qualidade do efluente do reator UASB em termos de DQO total e sólidos em suspensão, quando da alimentação desse reator com excesso de lodo aeróbio. Em 2016, a concentração média de DQO total no efluente foi da ordem de 529 mg/L e a concentração média de SST foi na ordem de 672 mg/L quando o afluente do reator UASB era composto com excesso de lodo aeróbio, ao passo que, sem excesso de lodo aeróbio no afluente, a DQO total média no efluente foi de 269 mg/L e a concentração média de SST foi de 349 mg/L. Porém, essa redução da qualidade pareceu estar relacionada a uma sobreposição de efeitos associado a presença de lodo aeróbio no afluente, e a sistemática adotada na estação para remoção de lodo anaeróbio do reator UASB. Quando não havia remoção de excesso de lodo anaeróbio de fundo do reator UASB, a qualidade do efluente se manteve durante a maior parte do tempo monitorado compatível com os períodos onde o afluente do reator UASB era composto essencialmente de esgoto bruto, indicando que o lodo de fundo, mais concentrado, exerce papel fundamental na retenção do lodo aeróbio na zona de digestão do reator UASB. Essa constatação se mostrou mais explícita quando as operações de descarte de excesso de lodo aeróbio e remoção de lodo anaeróbio do reator UASB ocorriam em períodos distintos. Nessa condição, a concentração média de DQO total no efluente foi de 254 mg/L e a concentração média de SST foi de 314 mg/L, ao passo que sem lodo aeróbio no afluente do reator UASB, a concentração média de DQO total no efluente foi de 191 mg/L e a concentração de SST foi de 258 mg/L. As características do biogás produzido variaram quando o sistema era alimentado com excesso de lodo aeróbio no afluente, apresentando, nessa condição, maior concentração de CO2 e menor concentração de H2S. Não houve, alteração da concentração média de metano quando da presença de excesso de lodo aeróbio no afluente. Outro aspecto observado foi a falta de correlação entre os parâmetros DQO removida e produção volumétrica de biogás quando o afluente do reator UASB continha excesso de lodo aeróbio. Na ausência de excesso de lodo aeróbio no afluente, houve correlação entre essas variáveis, porém uma correlação fraca, na ordem de 0,40, que pode ser explicada por uma captação ineficaz do biogás produzido, devido a vazamentos pela linha de quebra-escuma do separador trifásico, e por perdas de metano dissolvido com o efluente. Com relação ao impacto na unidade de pós-tratamento, a baixa capacidade de remoção de lodo de excesso da ETE devido a dificuldades de ordem operacional e contratual, condicionaram a operação do sistema de lodo ativado com elevadas concentrações de sólidos em suspensão. Nessa condição, o sistema de aeração apresentou baixa eficiência de transferência de oxigênio para o licor misto, o que impactou negativamente no consumo de energia elétrica do processo de lodo ativado, aumentando o custo operacional da ETE. A qualidade do efluente final da ETE, de modo geral, não foi impactada pela operação de descarte de lodo aeróbio no reator UASB, tendo mantido durante a maior parte do tempo DQO média na ordem de 40 mg/L no efluente e concentração de sólidos em suspensão totais em média na ordem de 25 mg/L. A nitrificação e desnitrificação foi variável ao longo do período do estudo, sendo que o processo apresentou eficiência média de remoção de nitrogênio em torno de 66%. / The objective of this research was to evaluate the effect of aerobic sludge discard on UASB reactor performance. In parallel, the effect of this operation on the final effluent quality of the WWTP, the operational performance of the post-treatment plant and the characteristics of the biogas produced in the UASB reactor were verified. In general, there was no negative impact on the anaerobic digestion process in terms of dissolved COD in the effluent, which was similar when the influent was composed essentially of raw sewage, in the order of 60 to 80 mg / L. On the other hand, the results indicated a tendency to reduce the quality of the UASB reactor effluent in terms of total COD and suspended solids when feeding this reactor also with excess aerobic sludge. In 2016, the mean concentration COD in the effluent was of the order of 529 mg / L and the mean concentration SST was 672 mg / L when the UASB reactor inflow was composed of excess aerobic sludge, while, without excess of aerobic sludge in the influent, the mean COD in the effluent was 269 mg / L and the mean concentration of SST was 349 mg / L. However, this quality reduction seemed to be related to an overlap of effects associated with the presence of aerobic sludge in the influent, and the system adopted at the WWTP to remove anaerobic sludge from the UASB reactor. When there was no removal of excess anaerobic bottom sludge from the UASB reactor, the quality of the effluent remained for most of the monitored time consistent with the periods where the UASB reactor inflow was essentially composed of raw sewage, indicating that the sludge from deep, more concentrated, plays a fundamental role in the retention of aerobic sludge in the digestion zone of the UASB reactor. This finding was more explicit when the operations of excess aerobic sludge and anaerobic sludge removal from the UASB reactor occurred at different periods. In this condition, the mean COD concentration in the effluent was 254 mg / L and the mean SST concentration was 314 mg / L, whereas without aerobic sludge in the UASB reactor, the mean total COD concentration in the effluent was 191 mg / L and the concentration of SST was 258 mg / L. The characteristics of the biogas produced varied when the system was fed with excess of aerobic sludge in the influent presenting, in this condition, a higher concentration of CO2 and a lower concentration of H2S. There was, fortunately, no change in the average concentration of methane when there was an excess of aerobic sludge in the influent. Another important aspect observed was the lack of correlation between the COD removal parameters and volumetric biogas production when the UASB reactor inflow contained excess aerobic sludge. In the absence of excess aerobic sludge in the influent, there was a correlation between these variables, but a weak correlation, in the order of 0.40, which can be explained by an inefficient capture of the biogas produced, due to leaks by the three-phase separator, and for losses of methane dissolved with the effluent. Regarding the impact on the post-treatment unit, the low sludge removal capacity of the WWTP due to operational and contractual difficulties, conditioned the operation of the activated sludge system with high concentrations of suspended solids. In this condition, the aeration system presented low oxygen transfer efficiency for the mixed liquor, which negatively impacted the electric power consumption of the activated sludge process, increasing the operational cost of the WWTP. The final effluent quality, in general, was not affected by the aerobic sludge discharge operation in the UASB reactor, having maintained for the most part the average COD in the order of 40 mg / L in the effluent and the solids concentration in suspension on average in the order of 25 mg / L. Nitrification and denitrification were variable throughout the study period, with the average nitrogen removal efficiency around 66%.

This study has examined the ten most shorted shares belonging to the Swedish Stockholm Stock Exchange's Large Cap list, by following randomly selected financial institutions that have chosen to take short positions. The purpose of the study is to investigate whether it is possible for short sellers to generate an excess return compared to the index OMXS30GI. The theory is mostly about short selling in general, efficient market hypothesis, behavioral finance, opponents of short selling, technical analysis of an index and the theory also includes previous research regarding short selling. The method used is based on collected secondary data from different databases. Via the secondary data, we have artificially followed randomly selected financial institutions that have glossed over and done the same as them to see if it can generate an excess return. In this study we will not take the cost associated with short selling into account which normally would be costs as margin interest, stock borrowing costs and commissions to brokers. The results show that it is possible for short sellers to generate an excess return that outperforms index OMXS30GI. We can also conclude that short sellers follow a pattern that indicates that they do not act in a way to destroy market efficiency and we can question whether the market is efficient or not.

Background: Investing in stocks seems to be more widespread than ever. The question is whether there are strategies that mean that investors systematically and over a longer period of time can generate excess returns. Proponents of the efficient market hypothesis believe that this is not possible as a higher return than the market is obtained only as a result of fortuity or a higher risk-taking. In contrast, there are those who believe that investment strategies, by exploiting market inefficiencies, can generate excess returns. This study examines whether this is possible by applying an investment strategy based on theoretically derived multiples. Purpose: The purpose of this study is to analyze whether an investment strategy based on theoretically derived multiples can generate excess returns by identifying mispricings in the Swedish stock market. Methodology: This study has been conducted with a quantitative research method and deductive approach to be able to achieve the purpose of the study. The study's measurement period extends from 2007-2022 and is limited to companies that have been listed on the OMX Stockholm Large Cap. During the measurement period, the investment strategy has been applied with three different investment horizons for the multiples P/E, P/BV, EV/EBITDA and EV/S. To examine the strategy's ability to generate excess returns, the return and risk of constructed portfolios have been analyzed in relation to OMXSPI. Result: The results of the study show that the strategy should be applied with a longer investment horizon. All portfolios with an investment horizon of one or three years have generated a higher return than the study's benchmark index. The strategy thus seems to be able to generate excess returns, even though the results indicate that it works better during periods of upturn in the market. As the proportion of shares developed in line with the forecast was over 50% for all multiples, the excess return also seems to be explained by the fact that the strategy works well for identifying mispricings.

Få studier är genomförda på området om hur kapitalförvaltare skapar och mäter överavkastning i praktiken, och ännu färre avseende svenska kapitalförvaltare. Det finns främst kvantitativa studier inom detta område vilket har lett till ett forskningsgap avseende kvalitativt inriktad forskning. Vidare är oroligt börsklimat en stor faktor för kapitalförvaltare att hantera, vilket innebär att de fortfarande behöver prestera under dessa perioder för att behålla sina kunder. Detta har lett till forskningsfrågan “Hur arbetar svenska kapitalförvaltare i praktiken för att skapa överavkastning och skiljer sig detta vid oroligt börsklimat?”. Kapitalförvaltare behöver också mäta prestationen för att kunna analysera den, men även för att visa befintliga och potentiella kunder tidigare resultat. Detta har resulterat i studiens andra forskningsfråga “Vilka prestationsmått används för att mäta överavkastning av svenska kapitalförvaltare och varför har de valt att använda dem?”. Denna studie är på grund av de två forskningsfrågorna både kvalitativ och kvantitativ. Detta åstadkoms genom att intervjua kapitalförvaltare men även genom en enkätundersökning för att besvara båda forskningsfrågorna. Sex kapitalförvaltare har deltagit i intervjuundersökningen och 37 kapitalförvaltare har svarat på enkätundersökningen. För enkätundersökningen motsvarar detta en svarsfrekvens om 38,5 % av de företag som kontaktats. Respondenterna i studien är anonyma och har tilldelats fiktiva namn, deras deltagande har dessutom varit frivilligt. Resultatet av denna studie visar att svenska kapitalförvaltare använder olika investeringsstrategier för att skapa överavkastning. Både fundamental och kvantitativ analys används, även om de används i kombination med varandra i olika utsträckning. Gällande ett oroligt börsklimat fokuserar respondenterna på att minska den tagna risken och att se till att vara väl diversifierad. För att mäta överavkastning är Sharpekvot och Informationskvot de mest frekvent använda prestationsmåtten i praktiken, i absoluta respektive relativa termer. Dessa har valts på grund av de är enkla och generellt sätt lätta att förstå även för mindre kunniga kunder. / Few studies have been made on the subject of how asset managers in practice create and measure performance and even fewer regarding Swedish asset managers. There are mainly quantitative studies made in this area, which has left a qualitative gap to research. Furthermore, a troubled stock market climate is a big factor for asset managers to deal with, which means that they still need to perform during these periods of time to keep their customers. This has led to the research question “How do Swedish asset managers work in practice to create excess return and does it differ during a troubled stock market climate?”. Asset managers also need to measure their performance to be able to analyze it, but also show existing and potential customers past results. This resulted in the study's second research question “Which performance measures are used to measure excess return by Swedish asset managers and why are they chosen?”.   This study is as a result of the two research questions both qualitative and quantitative. This is accomplished by having interviews with asset managers and also by performing a survey to answer both research questions. Six asset managers have participated in the interview survey and 37 asset managers have answered the survey. The survey has had a participation rate of 38,5 % of the companies that were contacted. The respondents in this study are anonymous and have been given fictional names, their participation have in addition been voluntarily.   The result of this study shows that Swedish asset managers are using different investment strategies to create excess return. Both fundamental and quantitative analysis are used, though these strategies are commonly used combined to varying extent. Regarding a troubled stock market climate, the respondents focus on lowering the risk taken and to make sure they are well diversified. To measure excess return Sharpe ratio and Information ratio are the most commonly used performance measures in practice, in terms of absolute and relative performance. These are chosen due to their simplicity and are overall easy to grasp even for less knowledgeable customers.

[pt] Este estudo objetiva investigar se existe uma relação entre posições agregadas de fundos de investimento brasileiros e o desempenho passado, presente e futuro das ações. Utilizando dados mensais das carteiras de 30.416 fundos entre 2006 e 2018 e dados de mercado de 84 ações, diversas regressões com dados em painel foram feitas para testar a correlação entre retornos totais das ações e o nível e a mudança de posições compradas e vendidas. Além disso, os dados foram utilizados para testar uma estratégia long-short de investimento segundo a qual as ações que ocupam posições altas nos fundos são compradas e as ações com baixas posições são vendidas. / [en] This study aims to investigate whether there is a relationship between aggregate Brazilian mutual fund positions and past, concurrent, and future stock performance. Using monthly fund portfolio data for 30.416 funds between 2006 and 2018 and market data for 84 stocks, several panel data regressions were ran to test the correlation between stock total returns and both the level and change in mutual fund aggregate long and short positions. In addition, the data was used in testing a long-short investing strategy in which stocks with high fund ownership indicators are purchased and stocks with low fund ownership indicators are sold.

Bakgrund: I takt med den stigande börsen så har intresset för aktiehandel ökat. Det finns ett stort antal investeringsstrategier att använda i jakten på att generera maximal avkastning. Två sådana strategier är att investera i företag med låga respektive höga multiplar. Tidigare studier har primärt fokuserat på icke teoretiska (redovisade) multiplar för stora bolag, varför det kan vara relevant att undersöka hur väl investeringsstrategier baserade på teoretiska multiplar för små och medelstora bolag presterar. Syfte: Syftet med denna studie är att med hjälp av teoretiska multiplar identifiera undervärderade bolag listade på Stockholmsbörsen Large, Mid samt Small Cap, för att sedan studera om en sådan strategi kan användas för att överavkasta över tid. Investeringarnas optimala tidshorisont undersöks också. Metod: Studien använder en kvantitativ forskningsstrategi med en deduktiv ansats. Bolagen som undersöks är listade på OMXSPI under perioden 2016–2022 med 3, 6, 9 och 12 månaders portföljer. De redovisade och teoretiska multiplarna P/E, EV/EBITDA och EV/Sales jämförs och totalt har nio portföljer skapats. Dessa portföljer utvärderas med avseende på årlig avkastning och ackumulerad avkastning samt med olika utvärderingsmått såsom Sharpekvot, Treynorkvot, Modigliani-Modigliani och Jensens alfa. Resultat: Resultatet visar att det inte är möjligt att överavkasta genom att investera i teoretiskt undervärderade bolag. Ingen av årsportföljerna hade en riskjusterad avkastning som översteg index. Den multipel som lyckades bäst i denna studie var EV/EBITDA, och den marknadslista som presterade bäst var Mid Cap, dock utan signifikans i resultaten vilket innebär att det kan ha varit slumpmässigt. Vad gäller portföljerna med kortare tidshorisont presterade de klart bättre än årsportföljerna, men även här är det svårt att dra några slutsatser eftersom signifikansen överlag var låg. / Background: With a stock market rising for a long time, interest in the share trading has increased. There is a great number of investment strategies to use in the pursuit of generating maximum returns. Two of these are to invest in companies with low and high multiples, respectively. Previous studies have primarily focused on non-theoretical (reported) multiples for large companies. Thus, it is relevant to examine how well investment strategies based on theoretical multiples will perform for small and medium companies. Purpose: The purpose of this study is to identify undervalued companies listed on the Stockholm Stock Exchange Large, Mid and Small Cap by means of theoretical multiples, and then study whether such a strategy can be used to over-return over time. The optimal time horizon for the investments is also examined. Method: In this study, we have used a quantitative research strategy with a deductive approach. The companies under investigation are listed on OMXSPI during the period 2016–2022 with 3, 6, 9 and 12- month portfolios. The reported and theoretical multiples P/E, EV/EBITDA, and EV/Sales are compared and a total of nine portfolios were created. These portfolios are evaluated with respect to annual returns and accumulated returns and with various evaluation measures such as Sharpe ratio, Treyone ratio, Modigliani-Modigliani, and Jensen's alpha. Result: The results of the study show that it is not possible to over-return by investing in theoretically undervalued companies. None of the annual portfolios managed, risk-adjusted, to outperform the market. The multiple that succeeded best in this study was EV/EBITDA, and the market list that performed best was Mid Cap, although without significance in the results, which means that it may have been random. As for the portfolios with a shorter time horizon, they performed clearly better than the 12-month portfolios, but even here it is difficult to draw any conclusions as the significance was generally low

近年來，波動度交易已成為熱門的研究議題。本研究探討貝他套利模式的理論及方法，並且利用台灣股票市場2003年到2012年的實證資料來研究BAB無系統風險套利模式的投資績效與超額報酬。我們買進低貝他投資組合並借券賣出高貝他投資組合，建立一個BAB權重的零成本投資組合。並且利用超額報酬與多因子alpha來衡量低貝他投資組合、高貝他投資組合及BAB套利方法的投資績效。此外，我們也探討融資限制對於BAB套利方法之報酬的影響。 本研究實證發現低貝他投資組合比高貝他投資組合在風險因子調整下的能賺取更高的超額報酬。而在台灣股票市場中，BAB套利模式在風險因子調整下更可以得到顯著的超額報酬；而融資限制也確實減低BAB套利模式的報酬率。 / This article examines the theories and methods of beta arbitrage, and considers the returns on market- neutral betting against beta (BAB) factors in Taiwan Stock Exchange (TSE) in during 2003 to 2012. We long low beta portfolios and short sell high beta portfolios, constructing a zero cost portfolio that weighted by BAB factor. Then we examine the excess return and multiple alphas for low beta portfolios, high beta portfolios, and BAB factor. Besides, we are interest that funding constraint might affect the BAB return. We find the evidence that low beta portfolios gain more return adjusted by risks than high beta portfolios. BAB factor could produce significant positive risk adjusted returns. Funding constraint might reduce the BAB return.