Contributions in Merlin

null

The energy demand of data centers has been steadily increasing over the years through increased usage of digital services. This thesis aims to elicitate information about the energy demand of the storage systems operated at UZH and moreover provides an estimate concerning the energy demand of a purely online-based university course. Theoretical background about the workings of a data center and the effect of COVID-19 on the educational environment are summarized in addition to proposed calculation models that determine the energy demand and emissions of storage systems at UZH and online lectures. The data gathered from the interviews with employees of UZH serves as the basis of the calculation model, while two video conferencing systems in use for online lectures were considered. In addition to the video conferencing systems, the case of online learnings' energy demand through prerecorded lectures has been estimated. The emission results indicate that the applied calculation model for the storage systems are viable, as they are within the range of the emissions that have been reported in the sustainability report 2020 of UZH. Organizational and technical measures to reduce the energy demand of the storage systems are furthermore proposed. The estimations for the online learning scenarios suggest that online learning may be an appropriate measure to reduce emissions caused by UZH. The quality of teaching and social aspects may in turn experience a negative effect.

With the seemingly increasing importance of net-zero pledges, this thesis aims to add to the existing literature by showing the effects of net-zero corporate climate strategies on company value. Based on existing literature, an increase in company value following a net-zero announcement due to lower costs, higher revenue, and lower climate risk is expected. The conducted analysis shows no significant results indicating a positive effect of net-zero announcements. In a second step, this thesis looks at the influence of governmental climate strategies on this effect of net-zero announcements on stock returns. Governments can influence people’s beliefs regarding climate change risk, and they also directly influence climate risk through climate regulations. The two presidencies of Trump and Biden are well suited for an investigation of different governmental climate strategies, as Trump’s election in 2016 drastically downshifted expectations about U.S. policy towards climate mitigation, and Biden’s election in 2020 had the opposite effect, drastically shifting them upward. In line with existing literature, a higher positive effect of the net-zero announcement on company value is expected under the presidency of Biden compared to Trump’s. No significant results are found when looking at the two presidencies. A possible reason for the insignificant results could be that investors do not consider net-zero announcements to be relevant or reliable.

Managers use item 7 (MD&A) of 10-Q and K filings to express their diagnosis of future plans, which can't be described in the form of numerical data. This paper examines the tone of sentiment of this qualitative data section of financial statements, by conducting a textual analysis. The tone of sentiment is used as an indicator for the predictive power of the S&P 500 financial statements on CAPM adjusted excess stock returns. SEC's EDGAR moved obligatory financial disclosures from the physical to the digital era, making such information available for everybody. This thesis finds that measuring tone could predict returns.

This thesis examines the shareholder wealth effect of spin-off announcements in the short and medium-term. I study the short-term abnormal returns based on a sample of 184 parent companies that announced a spin-off between 2000-2017. My findings indicate that spin-off announcements generate a positive cumulative adjusted average return of 4.15% within a twoday event window after the spin-off announcement. The abnormal returns are higher for larger spin-off announcements. Further, employing a Fama French three-factor model, I investigate the share performance of parent companies for up to 18 months after their spin-off announcement and find that, on average, parent companies generate a significant overperformance within the medium-term.

Over the last two decades, enormous losses associated with the momentum strategy surged again to an all-time high since the Great Depression. This thesis examines the contribution of the Constant and Dynamic Volatility Scaling Approach to the momentum effect in an industrial composition. Their contributions suggest that momentum risks are highly variable over time and partly predictable. I provide evidence that risk-scaling strategies can approximately double the Sharpe ratio of momentum in industries. Furthermore, improvements persist during various sub-sample periods. It appears that the Constant Volatility Approach is the most efficient strategy in this setting.

Marcos Lopez de Prado introduced meta-labeling, a unique machine-learning technique for improving algorithmic trading tactics, in 2018. Meta-labeling is a term that refers to the process of assigning a binary label to past trades of a trading system based on their outcome (win or loss), constructing a set of time series features that are temporally aligned with the labels, and fitting a machine learning classification model to the features and labels. The learned classifier is then used to evaluate the probability of profitability for each new, unopened trade. Before the trade is opened, the position size for each new transaction is computed using the corresponding probability estimate. The objective is to increase the position size of trades with a high projected probability of profitability. This thesis aims to determine whether or not meta-labeling improves trading performance by adjusting the bet size for each trade based on the probability of profitability predicted by secondary meta-labeled machine learning layers when applied to real-world trading systems and to draw critical conclusions about the concept’s practical implementation. Three different meta-labeled machine learning layers are built on top of three existing investment expertise, which could be fundamental or systematically driven. As a result, the three systems’ performance is enhanced. The findings indicate that meta-labeling really improves trading system performance and thus should be considered by all traders.

This paper examines if exposure towards climate risk reported in regulatory disclosures impacts asset returns. Doing so, it uses machine learning algorithms to quantify 8-K, 10-K as well as Management Call scripts. Training the algorithms to distinguish between opportunities as well as regulatory and physical climate risks, the thesis finds that an increased disclosure is positively associated with asset returns in the time-series. Also, opportunities and regulatory threats show an economically and statistically significant effect on the cross-sectional variation of asset returns, while physical exposure is only marginally priced. Based on these findings, the paper aims to further challenge standard approaches in empirical asset pricing that assume a sparse model configuration and faultless variable selection. Testing the results in a high-dimensional regression setting, a model-selection method that accounts for selection mistakes is proposed. This approach is argued to consistently evaluate the marginal variation of climate risk exposure relative to a large pool of factors. Consistent with earlier results, explanatory power is monotonically reduced for each specification due to the larger absorption of the underlying asset variation based on the covariate structure. However, opportunity-based exposure is shown to maintain statistical significance beyond the explanatory power of the factor pool.

null