Portfolio Optimization in DSE Using Financial Indicators, LSTM & PyportfolioOpt

Due to its suitable power to anticipate using Non-Linear forecasting methodologies, LSTM (Long Short-Term Memory) has changed the approach to time series prediction several folds. Process compatibilities of technical identifiers and various financial benchmarks that are defining financial decision-making in international markets are affecting Bangladesh Market as well. Issues like MACD and RSI as a technical investigator and financial ratio aspects of EPS and PE Ratio play an important role in the selection of assets in DSE. Given adequate training in line with intended functionality models, RNN has the potential to think through in a similar manner and the probable results are exhibited in this paper. Because of the Gated Structure, which refers to retaining important information and discarding irrelevant information through diminishing gradient and exploding gradient, LSTM has achieved significant advances in nonlinear forecasting that is based on human behavior. In this study, we compared two alternative portfolios that will be dependent on LSTM's future forecasting capabilities in terms of projecting the greatest potential output, which is demonstrated using Portfolio Optimization principles.

Chaos Theory Methods for Contact-Center Dinamic Control

The work is devoted to searching for optimal control methods for contact center, in particular, methods for predicting the load for further calculation of required number of operators. If the number of operators is always more than required, then the owners of the contact center will incur financial losses. If there are too few employees, the quality of service will decline. Predicting the load of the contact center is required in order to bring the optimal number of operators to work in advance. It is proposed to apply chaos theory to predict the incoming load of a contact center. Positive value of the Lyapunov index indicates the chaotic behavior of the input flow of the load. To predict the load, the methods of linear and nonlinear forecasting and the method of global approximation are used. The paper presents the results of comparing these methods for the problem of predicting the incoming load of contact center.

Improving the Accuracy and Time Interval of Predicting Ambient Parameters Applied to Dynamic Line Rating

This paper addresses wind speed prediction in the dynamic line rating (DLR) environment. We have described architecture of the DLR system as well as the main characteristics of nonlinear forecasting models, such as neural and fuzzy logic networks. Described models were tested and compared using real data (time series with data on wind speed, wind direction, air temperature, and solar radiation). The goal was to increase the accuracy and time of short-term prediction. The results show that neural networks outperform fuzzy logic and that the prediction time interval can be extended up to several hours, with no major compromise of the accuracy.

Nonlinear forecasting of Freight Volume based on Elman Network Model

Changes in the demand for bulk cargo will have a significant impact on the industrial structure and transportation development planning, and trains and cargo ships are the main means of bulk cargo transportation. Accurately predicting the volume of bulk cargo transportation can be used to support transportation system management, such as operation planning and route selection design. This study uses the Elman model to conduct empirical analysis, predict the changing trend of bulk cargo transportation, and provide data reference for the formulation of transportation development plans.

Nonlinear model of a dynamic system for forecasting of fuel consumption of a ship using a neural network

В современных условиях коммерческого судоходства, большое внимания уделяется вопросам оптимизации расхода топлива на судах. Наиболее критическим моментом, определяющим эффективность рейса, является количество бункерного топлива, использованного на морском переходе судном. В данной статье предложен подход к прогнозированию расхода топлива на предстоящем переходе судна с помощью использования нейронной сети, обученной с помощью алгоритма Левенберга-Марквардта, а также рассмотрено преимущество данного метода в сравнении с методами других исследователей. Статистическая выборка для машинного обучения составлена на основе эксплуатационных данных с танкера класса Афрамакс . Элементом новизны в данной работе является формирование данных для обучающего множества, а также возможность нелинейного прогнозирования посуточного приращения скорости. Данный метод имеет высокую точность и может применятся как фрахтователем, так и судоводителем для того, чтобы оценить экономическую эффективность предстающего рейса или выбрать оптимальный маршрут по параметру расхода топлива. Ещё одной задачей прогнозирования параметров судна на переходе с помощью нейронной сети является расчёт ожидаемых приращений скорости судна, что таблица расходов бункерного топлива может быть применена только при условиях не более 4-5 баллом во шкале Бофорта. In modern conditions of commercial shipping, much attention is paid to the optimization of fuel consumption on the sea. The most critical moment determining the voyages efficiency is the amount of bunker fuel used by the ship at the sea passage. This article proposes an approach to forecasting fuel consumption at the upcoming passage of a vessel using a neural network taught-in by the Levenberg-Marquardt algorithm, and also considers the advantage of this method in comparison with methods of other researchers. The statistical sample for machine learning is based on operational data from an Aframax class tanker. The novelty element in this work is the formation of data for the training set, as well as the possibility of nonlinear forecasting of the daily increment of speed. This method is highly accurate and can be used by both the charterer and the navigator in order to evaluate the economic efficiency of the upcoming voyage or to choose the optimal route according to the fuel consumption parameter. Another task of predicting the parameters of a vessel at a passage using a neural network is to calculate the expected increments of the vessels speed, with that the table of bunker fuel consumption can be applied only under conditions of no more than 4-5 points on the Beaufort scale.

Nonlinear Models Used to Analyze the Relation between Inflation and Unemployment

Many mathematical models have been developed in the last years in order to analyze economic phenomena and processes. Some of these models are optimization models, static or dynamic, while others are developed specially to study the evolution of economic phenomena. The topic of this paper is forecasting with nonlinear models. A few well-known nonlinear models are introduced, and their properties are discussed. The variety of nonlinear relationships is important both from the perspective of estimation and from the precision of forecasts in the medium and especially long term. Most nonlinear forecasting methods and all methods based on neural networks lead to predictions that have a better quality than the forecasts obtained by linear methods. The last section of this paper contains a detailed study of the relationship between inflation and unemployment and a numerical application with numerical data from Romania.

Frequently asked questions about nonlinear dynamics and empirical dynamic modelling

Complex nonlinear dynamics are ubiquitous in marine ecology. Empirical dynamic modelling can be used to infer ecosystem dynamics and species interactions while making minimal assumptions. Although there is growing enthusiasm for applying these methods, the background required to understand them is not typically part of contemporary marine ecology curricula, leading to numerous questions and potential misunderstanding. In this study, we provide a brief overview of empirical dynamic modelling, followed by answers to the ten most frequently asked questions about nonlinear dynamics and nonlinear forecasting.