published_or_final_version / Economics / Master / Master of Social Sciences

<p>This thesis investigates the profitability of the momentum strategy in the Swedish stock market. The momentum strategy is an investment strategy where past winners are bought and past losers are sold short. In this paper Swedish stocks are analyzed during the period 1999 – 2007 with the approach first used by Jegadeesh and Titman (1993). The results indicate that momentum investing is profitable on the Swedish market. The main contribution to the profits is derived from investing in winners while the losers in most cases do not contribute at all to total profits. The profits remain after correcting for transaction costs for longer termed strategies while they diminish for the shorter termed ones. Compared to the market index, buying past winners yield an excess return while short selling of losers tend to make index investing more profitable. The analysis also shows that momentum can not be explained by the systematic risk of the individual stocks. The evidence in support of a momentum effect presented in this thesis also implies that predictable price patterns can be used to make excess returns; this contradicts the efficient market hypothesis.</p>

Oestrogen decline during the menopause leads to decline in cognitive performance because oestrogen receptor sites are found in the prefrontal cortex and hippocampus of the female brain, areas associated with memory and attention functions. Extensive research over the past two decades has tested the effects of administering Hormone Replacement Therapy (HRT) to maintain oestrogen levels. MRI studies have shown improvements in hippocampal volume and frontal functions with HRT, but evidence for associated improvements in verbal memory performance has been mixed. Some studies have even found detrimental effects of HRT, leading to the suggestion of a critical period for HRT administration relative to menopause. Oestrogen receptor sites are found in frontal brain regions associated with working memory (WM) functions including attention. These functions have been researched less than verbal memory, but with similarly mixed findings. The research reported in this thesis tested the critical period hypothesis in relation to WM. Study one tested the prediction that HRT will benefit WM if the therapy is initiated during the peri-menopause, and will harm it if initiated post-menopause. A naturalistic sample of 121 women were recruited, comprising women who varied in the time they had begun taking HRT, and menopausal status-matched controls who had never taken HRT. Participants completed three tests of WM span and the Sustained Attention to Response Task (SART) on two occasions 12 months apart. WM performance supported the critical period hypothesis, with women who had begun the therapy after the menopause displayed worsened WM capacity when compared to peri-menopausal initiators and post-menopausal women with no history of HRT use. At one year follow up, postmenopausal HRT users were still underperforming compared to peri-HRT initiators and those in the post-menopausal stage with no history of HRT use. No significant differences were identified between groups on the SART. The effects of natural supplements on physical symptoms of the menopause have been researched, but there is little research on their effects on cognitive symptoms and none specifically testing the critical period hypothesis. Study two tested the effects of soya isoflavones on WM during peri- and post-menopausal stages. One hundred and twelve peri- and post-menopausal women were randomly allocated to receive either placebo or 100mg soya supplement in capsules daily for three months. Participants and researcher were blind to this allocation. Participants completed two tests of WM span and two Sustained Attention (SA) tasks at baseline, after three months of soya/placebo, and after a further three months without supplement. There was no effect of isoflavones on cognition, regardless of time of initiation of the supplement. This thesis offers a unique contribution to the literature, by establishing empirically that HRT may have long-lasting benefits for WM if administered in the peri-menopause period, and detriments if taken post-menopause. There was no evidence that administration of soya-based phytoestrogens for three months peri- or post-menopause replicated these effects of HRT on cognition.

The significant impact of method of payment on the share price abnormal returns following mergers and acquisitions have been broadly considered and documented in US and UK empirical studies (Agrawal and Jaffe 2000). In the UK, all-cash acquisitions show insignificant negative or small positive abnormal returns, whilst the all-equity acquisitions have significant negative returns. Whilst it is tempting to conclude that it is simply the form of financing that separates the shareholder value destruction of equity-financed takeovers from cash takeovers, such a conclusion tends to ignore the question of where the cash to fund the acquisition comes from in the first place. Theory tells us this should matter. Whilst different theories on firm financing offer competing explanations on both managerial choices and shareholder preferences, it seems reasonable to ask the question whether the source of the cash influences the long run wealth effect of the acquisition. In order to shed light on this issue, this investigation looks at short-term daily abnormal returns as well as long-term abnormal returns including a five-year horizon of post-takeover returns and a three-year horizon of pre-takeover returns. The short-term daily abnormal returns support the signalling information hypothesis to some extent as acquirers financing takeovers using internal cash out-perform those financing takeovers by equity or debt issues. After categorizing the research sample firms into two sub-groups, one being internal funding while the other being external sources including equity or debt, the share price abnormal returns show statistically significant differences between these two sub-groups over 11-day event windows. Further, by using one- and two-dimensional analyses and a univariate test, the results reveal that UK cash acquisitions explored by this investigation contradict the free cash flow (FCF) hypothesis. Regression models show that book-to-market ratio is important in explaining the short-term daily abnormal returns. The long-term post-takeover stock performances show sensitivity to the benchmark adopted as well as the calculation used for the long-term abnormal returns, i.e. cumulated or compounded. Owing to the small sample firms entering the calendar time monthly portfolios, the calendar time approach employs White (1980) corrections and a GLS model to mitigate the effects of heteroskedasticity in the research sample. Generally speaking, long-term abnormal returns show a negative pattern for the whole sample as well as the sub-groups depending on their dominant financing methods. Furthermore, the univariate and multivariate tests demonstrate that the FCF hypothesis cannot explain the 60-month share price abnormal returns of the research sample. According to the coefficient derived from regression model(s), the most significant factor to predict 60-month abnormal returns is relative size (market value of target to that of bidder). The results suggest that the bigger the relative size of the target, the more negative the abnormal return will be (Hansen 1987, Martin 1996, Loughran and Vijh 1997). Besides, the institutional investors contribute a positive effect on long-term share price performance, which is consistent with the findings of Chen, Harford, and Li (2007). The pre-takeover share price abnormal returns over three years intervals prior to the bid announcements clearly show that cash acquirers overall experience a significant positive stock performance. This result is robust to adopting various benchmarks of event time and calendar time regression-based framework. Based on the dominant financing method used for the acquirers, firms issuing debt before the bid announcements do perform extremely well. Those firms subsequently perform badly for post-takeover long-term intervals. Accordingly, this phenomenon demonstrates a mean reversion picture. Regardless of whether an event time or a calendar time approach is used, high q firms always have higher abnormal returns even when allowing for other factors, such as free cash flow or cash stock. However, multinomial logistic tests fail to find any statistically significant link between pre- takeover abnormal returns and the form of financing.

The importance of a well-functioning housing market has been proposed for long within economics, economic geography and urban planning.  A high mobility on the housing market most likely positively affects the dynamics of the labor market, a dynamic important for economic growth. Mobility defined as the link between the worker and the workplace in terms of transportation and housing are most likely essential components of a dynamic and well-functioning labor market. The Oswald hypothesis states that positive relationship between homeownership and unemployment exists, the lower mobility in the homeownership housing stock compared to the rental housing stock affects labor market mobility in a negative way which can be noted if European countries are compared. My thesis explores this relationship in a Swedish context by mobilizing a quantitative approach with aggregate data on municipal level ranging from 1998 to 2013. The Swedish housing market is in a deregulation process since 1992, a conversion process from public rental housing to homeownership co-op apartments has taken place and public policies now favor homeownership over renting. Municipal data on unemployment, homeownership of apartment, rental tenant and control variables for economy and personal characteristics are applied in Pooled OLS, random effects and fixed effects regression models. The results from the Pooled OLS and the Random effects model confirms the positive relationship proposed by Oswald for homeownership of apartment but not for homeownership of detached housing. Also rental tenant show a positive relationship. The results from the fixed effect estimation rejects the hypothesis altogether and show a negative relationship.

This study aims to describe how researchers may incorporate four different generalisation methods (working hypothesis, analytic generalisation, critical case and naturalistic generalisation) to aid them in properly generaliseing from case study results. It was found that critical cases can be used in combination with working hypothesis to determine the falsifiability of the hypotheses. Working hypothesis and analytic generalisation complement each other by making it possible for researchers to generalise to both theory and new hypotheses (i.e. both inductively and deductively). Lastly, naturalistic and analytic generalisation enables a double-ended generalisation, where both the reader and the scientist themself generalise to specific situations and over-all theory, respectively.

The purpose of this research was to evaluate the effect on parametric and nonparametric tests using ordinal data when the distances between points changed on the measurement scale. The research examined the performance of Type I and Type II error rates using selected parametric and nonparametric tests.

The assessment of the performance of learners by means of benchmark experiments is established exercise. In practice, benchmark studies are a tool to compare the performance of several competing algorithms for a certain learning problem. Cross-validation or resampling techniques are commonly used to derive point estimates of the performances which are compared to identify algorithms with good properties. For several benchmarking problems, test procedures taking the variability of those point estimates into account have been suggested. Most of the recently proposed inference procedures are based on special variance estimators for the cross-validated performance. We introduce a theoretical framework for inference problems in benchmark experiments and show that standard statistical test procedures can be used to test for differences in the performances. The theory is based on well defined distributions of performance measures which can be compared with established tests. To demonstrate the usefulness in practice, the theoretical results are applied to benchmark studies in a supervised learning situation based on artificial and real-world data. / Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"

This paper is a study on some upper bounds of the Mertens function, which is often considered somewhat of a ``mysterious" function in mathematics and is closely related to the Riemann Hypothesis. We discuss some known bounds of the Mertens function, and also seek new bounds with the help of an automated conjecture-making program named CONJECTURING, which was created by C. Larson and N. Van Cleemput, and inspired by Fajtowicz's Dalmatian Heuristic. By utilizing this powerful program, we were able to form, validate, and disprove hypotheses regarding the Mertens function and how it is bounded.

Doctor of Philosophy / Department of Statistics / Gary L. Gadbury / Multiple testing research has undergone renewed focus in recent years as advances in high throughput technologies have produced data on unprecedented scales. Much of the focus has been on false discovery rates (FDR) and related quantities that are estimated (or controlled for) in large scale multiple testing situations. Recent papers by Efron have directly addressed this issue and incorporated measures to account for high-dimensional correlation structure when estimating false discovery rates and when estimating a density. Other authors also have proposed methods to control or estimate FDR under dependencies with certain assumptions. However, not much focus is given to the stability of the results obtained under dependencies in the literature. This work begins by demonstrating the effect of dependence structure on the variance of the number of discoveries and the false discovery proportion (FDP). A variance of the number of discoveries is shown and the density of a test statistic, conditioned on the status (reject or failure to reject) of a different correlated test, is derived. A closed form solution to the correlation between test statistics is also derived. This correlation is a combination of correlations and variances of the data within groups being compared. It is shown that these correlations among the test statistics affect the conditional density and alters the threshold for significance of a correlated test, causing instability in the results. The concept of performing tests within networks, Conditional Network Testing (CNT) is introduced. This method is based on the conditional density mentioned above and uses the correlation between test statistics to construct networks. A method to simulate realistic data with preserved dependence structures is also presented. CNT is evaluated using simple simulations and the proposed simulation method. In addition, existing methods that controls false discovery rates are used on t-tests and CNT for comparing performance. It was shown that the false discovery proportion and type I error proportions are smaller when using CNT versus using t-tests and, in general, results are more stable when applied to CNT. Finally, applications and steps to further improve CNT are discussed.