Terms Of Annualized Return Research Articles

Predictability of commodity futures returns with machine learning models

AbstractWe use prevailing machine learning models to investigate the predictability of futures returns in 22 commodities with commodity‐specific and macroeconomic factors as predictors. Out‐of‐sample prediction errors for the majority of futures contracts are lowered compared with those obtained by the baseline models of AR(1) and forecast combinations. Using Shapley values to explain feature importance, we identify dominant predictors for each commodity. A long–short portfolio strategy based on monthly light gradient‐boosting machine predictions outperforms the benchmark linear models in terms of annual return, Sharpe ratio, and max drawdown.

A case study for the evaluation of realistic energy retrofit strategies for public office buildings in the Southern Hemisphere

The building sector offers significant opportunities for reducing the energy consumption with considerable economic, environmental and health benefits. Governments can lead the way by retrofitting existing public buildings to reinforce their commitment to improve energy efficiency. Similar design standards, end-uses and operational profiles are usually established for public buildings based on the services they offer. Retrofitting a public building can therefore serve as an ideal test-bed for energy efficiency measures for other buildings within a particular service category. This study first analysed the current electricity consumption of a public office building in Mauritius, located in the Southern Hemisphere. A complete model of the building was created, validated and then simulated to investigate the impact of realistic retrofit strategies on the electricity consumption. Results showed that lighting retrofit achieved the most significant reduction while measures that improved the thermal envelope of the building resulted in smaller energy savings. The possibility of exploiting solar energy was explored by simulating a 70 kWp photovoltaic system installed on the roof. An equivalent of 8.5% of the annual electricity consumption of the building could thus be generated. A financial analysis is also presented for all retrofit scenarios in terms of annual return and payback period.

New technologies for an effective energy retrofit of hospitals

Following the approval by the European Parliament of the directive on energy efficiency of public buildings, a great effort has been directed towards enhancing low-emission systems such as fuel cells (FC), photovoltaic systems (PVS) or solar thermal systems (STS), especially in all public facilities, such as hospitals, characterized by relevant energy requirements. This paper develops a theoretical analysis which focuses on the environmental benefits achievable through a shift from the conventional systems, normally operating in hospitals, to various hybrid plants. The model site is a hospital located near Ferrara (Italy). Several hybrid schemes were investigated and compared: PAFCs (phosporic acid fuel cells), STS and PVS. An energy analysis was developed for each option assuming, as a benchmark, the conventional systems operating today in the medical center. The results, presented with reference to the primary energy requirements and the pollutant emissions, demonstrate that in the case of existing systems being upgraded with these hybrid plants, overall emissions could be abated with a significant reduction in fossil energy consumption. Finally, an economic study, even taking external factors into account, is developed for all the retrofit scenarios in terms of annual return, simple payback period and IRR.