Historical Time Series Data Research Articles

A novel approach to incorporate investor’s preference in fuzzy multi-objective portfolio selection problem using credibility measure

Investment decision making is usually a multi-objective optimization problem in an uncertain environment. In a real-life scenario, an investor aims to choose a portfolio based on his/her preferences for each objective. Credibility theory has been attributed as one of the most effective ways to model uncertain portfolio attributes. LR fuzzy numbers based credibilistic moments have been used successfully in the recent past to model portfolio selection problems in uncertain environments. Traditionally, the preferences on portfolio attributes are reflected as user defined threshold values. This work proposes a novel way to integrate the investor’s preferences in the portfolio selection model. Investor’s preferences are specified as LR fuzzy numbers corresponding to the portfolio’s expected return and illiquidity distributions. The portfolio models formulated in this study attempt to attain them to the best possible extent. Four new models are presented considering uncertain portfolio attributes, namely, fuzzy return and fuzzy illiquidity. The models are solved using the MIBEX-SM Genetic Algorithm (GA) and ECE methodology. In order to demonstrate the efficient working of the proposed models, practical applications of portfolio selection problem using historical time series data of National Stock Exchange (NSE) is presented. Credibilistic Sharpe Ratio (CSR) has been used to compare the efficiency of models among themselves as well as with the Nifty−50 index.

Forecasting of Total Column Ozone using Regression Analysis and LSTM-RNN Machine Learning Approach

Objectives: Atmospheric Ozone plays an important role in global climate change, human health and environmental conditions. Accurate and timely prediction of variation in Total Column Ozone (TCO) concentration is very important for both climatology and environment. In this context, the present study aims to utilize the advancement of Artificial Intelligence in Machine learning and a conventional method to develop models for the prediction of TCO concentration using historical time series data over a tropical region, India. Methods: Long Short-Term Memory (LSTM) deep learning networks are very useful in classifying, processing and making predictions based on time series data. In this work, Multiple Linear Regression (MLR) model, Long Short-Term Memory (LSTM) models are developed for forecasting TCO over a tropical region. The Statistical Parameters such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) were used to analyze and evaluate the performance of the proposed models. Findings: In the present study the TCO concentration varied between 217.4 DU and 288.6 DU. The performances of MLR and LSTM model were compared. From the results, it is evident that the values of RMSE, MAE and MAPE of the MLR model are 3.807924, 2.93766, 1.15 respectively whereas the values for LSTM model are 3.074492, 2.574, 1.01 respectively. LSTM was found to be the accurate model for the study data set and the predicted TCO concentration has a very good correlation with the actual observations and it does not rely on the meteorological parameters. Novelty: The LSTM models are usually employed in the field of machine translation, speech recognition and for the prediction of traffic levels, stock levels and air pollutants. This work utilizes the fast numerical machine learning computing library Tensorflow and a high level neural network library Keras that runs on top of Tensorflow to develop a LSTM network model for the prediction of the concentration levels of TCO. Keywords: Total Column Ozone; Regression; Forecasting; Climate Change; Long Short-Term Memory

A Hybrid ARIMA-GABP Model for Predicting Sea Surface Temperature

Sea surface temperature (SST) is one of the most important parameters in air–sea interaction, and its accurate prediction is of great significance in the study of global climate change. However, SST is affected by heat flux, ocean dynamic processes, cloud coverage, and other factors, which means it contains linear and nonlinear components. Existing prediction models, especially single prediction models, cannot effectively handle these linear and nonlinear components in the meantime, degrading their accuracy concerning the prediction of SST. To remedy this weakness, this paper proposes a novel prediction model by the Lagrange multiplier method to combine the auto-regressive integrated moving average (ARIMA) model and the back propagation (BP) neural network model, where these two models have superior prediction performance for linear and nonlinear components, respectively. Moreover, the genetic algorithm is exploited to construct the genetic algorithm BP (GABP) neural network to further improve the performance of the proposed model. To verify the effectiveness of the proposed model, experiments predicting the SST based on historic time-series data are performed. The experiment results indicate that the mean absolute error (MAE) of the ARIMA-GABP model is only 0.3033 °C and the root mean square error (RMSE) is 0.3970 °C, which is better than the ARIMA model, BP neural network model, long short-term memory (LSTM) model, GABP neural network model, and ensemble empirical model decomposition BP model among various datasets. Therefore, the proposed model has superior and robust performance concerning predicting SST.

THE ROLE ASYMMETRIES IN EXCHANGE RATE REGIMES AND MACROECONOMIC PERFORMANCE: THE CASE OF NIGERIA

The role of exchange rate in the macroeconomic fundamentals of a country cannot be underestimated but the nonlinear (asymmetric) feature of the exchange rate movements becomes of importance when analyzing its role in macroeconomic performance. Thus, this study, using historical annual time series data from 1970 and 2020, and a nonlinear ARDL model investigated the nexus between exchange rate regimes and the macroeconomic performance in Nigeria. This model allows us to capture that the partial sum of positive exchange rate movements (exchange rate depreciations) and negative exchange rate movements (exchange appreciations). We found the potential exchange rate regime to cause declining inflationary pressure sensitive to whether the exchange rate regime is responding to depreciation or appreciation in its movement. But then, irrespective of whether the exchange rate is depreciating or appreciation, the exchange rate regimes exhibit little or no significant impact on output growth in Nigeria. Specifically, we find that an intermediate exchange rate regime based on exchange rate depreciations is viable for promoting trade surpluses, whereas intermediate exchange rate regime that is due to appreciation of exchange rate tends to cause to trade deficit. More importantly, we found that the magnitude of the role of the nonlinear feature of exchange rate on trade balance is relatively higher when the exchange rate is appreciating.

Predicting Energy Consumption Using LSTM, Multi-Layer GRU and Drop-GRU Neural Networks.

With the steep rise in the development of smart grids and the current advancement in developing measuring infrastructure, short term power consumption forecasting has recently gained increasing attention. In fact, the prediction of future power loads turns out to be a key issue to avoid energy wastage and to build effective power management strategies. Furthermore, energy consumption information can be considered historical time series data that are required to extract all meaningful knowledge and then forecast the future consumption. In this work, we aim to model and to compare three different machine learning algorithms in making a time series power forecast. The proposed models are the Long Short-Term Memory (LSTM), the Gated Recurrent Unit (GRU) and the Drop-GRU. We are going to use the power consumption data as our time series dataset and make predictions accordingly. The LSTM neural network has been favored in this work to predict the future load consumption and prevent consumption peaks. To provide a comprehensive evaluation of this method, we have performed several experiments using real data power consumption in some French cities. Experimental results on various time horizons show that the LSTM model produces a better result than the GRU and the Drop-GRU forecasting methods. There are fewer prediction errors and its precision is finer. Therefore, these predictions based on the LSTM method will allow us to make decisions in advance and trigger load shedding in cases where consumption exceeds the authorized threshold. This will have a significant impact on planning the power quality and the maintenance of power equipment.

Forecasting Carbon Price in China: A Multimodel Comparison.

With the global concern for carbon dioxide, the carbon emission trading market is becoming more and more important. An accurate forecast of carbon price plays a significant role in understanding the dynamics of the carbon trading market and achieving national emission reduction targets. Carbon prices are influenced by many factors, which makes carbon price forecasting a complicated problem. In recent years, deep learning models are widely used in price forecasting, because they have high forecasting accuracy when dealing with nonlinear time series data. In this paper, Multivariate Long Short-Term Memory (LSTM) in deep learning is used to forecast carbon prices in China, which takes into account the factors affecting the carbon price. The historical time series data of carbon prices in Hubei (HBEA) and Guangdong (GDEA) and three traditional energy prices affecting carbon prices from 5 May 2014 to 22 July 2021 are collected to form two data sets. To prove the forecast effect of our model, this paper not only uses Multivariate LSTM, Multilayer Perceptron (MLP), Support Vector Regression (SVR), and Recurrent Neural Network (RNN) to forecast the same data, but also compares the forecast results of Multivariate LSTM with the existing research on HBEA and GDEA forecast based on deep learning recently. The results show that the MAE, MSE, and RMSE obtained by the Multivariate LSTM are all smaller than other prediction models, which proves that the model is more suitable for carbon price forecast and offers a new approach to carbon prices forecast. This research conclusion also provides some policy implications.

Hourly Power Consumption Forecasting Using RobustSTL and TCN

Power consumption forecasting is a crucial need for power management to achieve sustainable energy. The power demand is increasing over time, while the forecasting of power consumption possesses challenges with nonlinearity patterns and various noise in the datasets. To this end, this paper proposes the RobustSTL and temporal convolutional network (TCN) model to forecast hourly power consumption. Through the RobustSTL, instead of standard STL, this decomposition method can extract time series data despite containing dynamic patterns, various noise, and burstiness. The trend, seasonality, and remainder components obtained from the decomposition operation can enhance prediction accuracy by providing significant information from the dataset. These components are then used as input for the TCN model applying deep learning for forecasting. TCN employing dilated causal convolutions and residual blocks to extract long-term data patterns outperforms recurrent networks in time series forecasting studies. To assess the proposed model, this paper conducts a comparison experiment between the proposed model and counterpart models. The result shows that the proposed model can grasp the rules of historical time series data related to hourly power consumption. Our proposed model overcomes the counterpart schemes in MAPE, MAE, and RMSE metrics. Additionally, the proposed model obtains the best results in precision, recall, and F1-score values. The result also indicates that the predicted data can fit the pattern of the actual data.

Probabilistic time series forecasting with deep non‐linear state space models

AbstractProbabilistic time series forecasting aims at estimating future probabilistic distributions based on given time series observations. It is a widespread challenge in various tasks, such as risk management and decision making. To investigate temporal patterns in time series data and predict subsequent probabilities, the state space model (SSM) provides a general framework. Variants of SSM achieve considerable success in many fields, such as engineering and statistics. However, since underlying processes in real‐world scenarios are usually unknown and complicated, actual time series observations are always irregular and noisy. Therefore, it is very difficult to determinate an SSM for classical statistical approaches. In this paper, a general time series forecasting framework, called Deep Non‐linear State Space Model (DNLSSM), is proposed to predict the probabilistic distribution based on estimated underlying unknown processes from historical time series data. We fuse deep neural networks and statistical methods to iteratively estimate states and network parameters and thus exploit intricate temporal patterns of time series data. In particular, the unscented Kalman filter (UKF) is adopted to calculate marginal likelihoods and update distributions recursively for non‐linear functions. After that, a non‐linear Joseph form covariance update is developed to ensure that calculated covariance matrices in UKF updates are symmetric and positive definitive. Therefore, the authors enhance the tolerance of UKF to round‐off errors and manage to combine UKF and deep neural networks. In this manner, the DNLSSM effectively models non‐linear correlations between observed time series data and underlying dynamic processes. Experiments in both synthetic and real‐world datasets demonstrate that the DNLSSM consistently improves the accuracy of probability forecasts compared to the baseline methods.

Spatio-temporal variability of the zooplankton community in the SW Mediterranean 1992–2020: Linkages with environmental drivers

Variability in the spatial and temporal distribution of the mesozooplankton abundance in the N Alboran Sea (SW Mediterranean) was assessed intermittently from 2010 to 2020, and compared with 1992–2000 historical time series data. Total abundance of mesozooplankton was significantly higher in the coast than in the shelf and slope waters. There were significant differences in mesozooplankton abundance between 1992–2000 and 2010–2020 at the three zones. Copepods dominated the mesozooplankton during winter and spring, but cladocerans and doliolids also became important components of the community in summer and autumn. We found significant increases between the first and the second decadal periods in the abundance of copepods, appendicularians, holoplanktonic gastropods and siphonophores in the shelf. However, in the coast, copepod nauplii, doliolids, gastropods and siphonophores increased, while euphausiids abundance decreased significantly. These trends contrast with the ongoing decline of the sardine stocks in European waters. Increasing temperature and decreasing predation pressure are suggested to be the main drivers of mesozooplankton variability.

Measuring national mood with music: using machine learning to construct a measure of national valence from audio data

We propose a new measure of national valence based on the emotional content of a country’s most popular songs. We first trained a machine learning model using 191 different audio features embedded within music and use this model to construct a long-run valence index for the UK. This index correlates strongly and significantly with survey-based life satisfaction and outperforms an equivalent text-based measure. Our methods have the potential to be applied widely and to provide a solution to the severe lack of historical time-series data on psychological well-being.

The role of R&D and economic policy uncertainty in Sri Lanka\u2019s economic growth

In this paper, we assess the role of investment in research and development (R&D) and economic policy uncertainty (EPU) in Sri Lanka’s economic growth experience. We do this by first determining which endogenous growth theories best explain the evolution of total factor productivity (TFP) in the country. Using historical time series data (1980–2018), we find that semi-endogenous growth theories best explain the evolution of TFP in Sri Lanka. This evidence suggests that R&D is critical to the country’s TFP expansion. We find that, through R&D, EPU has a crucial detrimental impact on TFP growth, although it is short-lived. Our findings are robust and have important implications for R&D investment and for moderating EPU.

Prediction of COVID-19 Transmission in the United States Using Google Search Trends

Accurate forecasting of emerging infectious diseases can guide public health officials in making appropriate decisions related to the allocation of public health resources. Due to the exponential spread of the COVID-19 infection worldwide, several computational models for forecasting the transmission and mortality rates of COVID-19 have been proposed in the literature. To accelerate scientific and public health insights into the spread and impact of COVID-19, Google released the Google COVID-19 search trends symptoms open-access dataset. Our objective is to develop 7 and 14-day-ahead forecasting models of COVID-19 transmission and mortality in the US using the Google search trends for COVID-19 related symptoms. Specifically, we propose a stacked long short-term memory (SLSTM) architecture for predicting COVID-19 confirmed and death cases using historical time series data combined with auxiliary time series data from the Google COVID-19 search trends symptoms dataset. Considering the SLSTM networks trained using historical data only as the base models, our base models for 7 and 14-day-ahead forecasting of COVID cases had the mean absolute percentage error (MAPE) values of 6.6% and 8.8%, respectively. On the other side, our proposed models had improved MAPE values of 3.2% and 5.6%, respectively. For 7 and 14 -day-ahead forecasting of COVID-19 deaths, the MAPE values of the base models were 4.8% and 11.4%, while the improved MAPE values of our proposed models were 4.7% and 7.8%, respectively. We found that the Google search trends for “pneumonia,” “shortness of breath,” and “fever” are the most informative search trends for predicting COVID-19 transmission. We also found that the search trends for “hypoxia” and “fever” were the most informative trends for forecasting COVID-19 mortality.

Deep Learning Framework for Physical Internet Hubs Inbound Containers Forecasting

This article presents a framework for physical internet hubs inbound containers forecasting based on deep learning and time series analysis. The inbound containers forecasting is essential for planning, scheduling, and resources allocation. The proposed framework consists of three main phases. First, the inbound historical transaction has been processed to find out the training window size (lags) using auto correlation function (ACF) and partial autocorrelation function (PACF). Second, the framework uses convolutional neural network (CNN) and recurrent neural network (RNN) as training networks for the historical time series data in two techniques. The proposed framework uses univariate and multivariate time series analysis to explore the maximum forecasting outcomes. Last, the framework measures the accuracy and compares the forecasting output using mean absolute error matrix (MAE) for both approaches. The experiments illustrated that RNN forecasts univariate inbound transaction with total 5.0954 MAE rather than 5.0236 for CNN. The CNN outperforms multivariate inbound containers forecasting with 0.7978 MAE. All the results has been compared with autoregressive integrated moving average (ARIMA) and support vector machine (SVR).

Digital Twin for Inventory Planning of Fresh Produce

The management of perishable food inventory demands special attention. Fruits quickly lose their freshness and perish if they are not consumed within a specified period. It is critical to develop a management tool based on the Internet of Things that can efficiently integrate all the dynamic data associated with various types of resources in real-time along the supply chain. This research is part of a comprehensive supply chain framework developed to analyze food bank logistics supply chain interactions. The study will mainly focus on the use of historical time-series data to create a digital twin that can anticipate future events. The digital twin framework was built based on the operational trend of the Italian food bank to strengthen the decision support system related to the fresh food inventory. The SAP Analytics Cloud was used to create a solution that would help the organization better satisfy consumer needs by reducing fruit waste in the inventory.

A Study on Pork Price Prediction Using LSTM

In this study, we propose LSTM, a technique for predicting pork prices to stabilize the supply and demand of pork. The LSTM model is trained using the attributes of day and month as additional inputs to historical time series data. The performance of the proposed LSTM model was compared to that of ARIMA, SARIMA, autoARIMA, and standard LSTM in terms of three performance metrics: mean absolute error (MAE), root mean squared error (RMSE), and mean absolute percentage error (MAPE). As a result of the experiment, the error rate of the proposed LSTM model was significantly lower than that of other existing methods. In future research, it will be necessary to redefine the analysis of causal relationships such as livestock substitutes and supplements, and to conduct practical studies such as linking OpenAPI with big data of public institutions to improve the accuracy of price prediction.

Agricultural supply response and price risk of maize and sorghum in South Africa

The study used the Autoregressive Distributed Lag-Error Correction Model (ARDL-ECM) approach to estimate the responsiveness of South African maize and sorghum producers to price risk, price incentives and non-price incentives. The price risk variable was incorporated in the supply models to examine its impact on maize and sorghum production decisions. The study used annual historical time series data of 49 observations for the period 1970 to 2018 was used in the analysis. The empirical results reveal that maize and sorghum producers' response to own price is reasonably low. The study further shows that both maize and sorghum crops demonstrate a high speed of adjustment to the long-run equilibrium, which means that in the event of a shock to the system, grain output will quickly re-establish itself at a faster rate. The findings underscore the relevance of price risk in determining production output in South Africa.

Stock Price Movement Prediction Using Sentiment Analysis and CandleStick Chart Representation.

Determining the price movement of stocks is a challenging problem to solve because of factors such as industry performance, economic variables, investor sentiment, company news, company performance, and social media sentiment. People can predict the price movement of stocks by applying machine learning algorithms on information contained in historical data, stock candlestick-chart data, and social-media data. However, it is hard to predict stock movement based on a single classifier. In this study, we proposed a multichannel collaborative network by incorporating candlestick-chart and social-media data for stock trend predictions. We first extracted the social media sentiment features using the Natural Language Toolkit and sentiment analysis data from Twitter. We then transformed the stock’s historical time series data into a candlestick chart to elucidate patterns in the stock’s movement. Finally, we integrated the stock’s sentiment features and its candlestick chart to predict the stock price movement over 4-, 6-, 8-, and 10-day time periods. Our collaborative network consisted of two branches: the first branch contained a one-dimensional convolutional neural network (CNN) performing sentiment classification. The second branch included a two-dimensional (2D) CNN performing image classifications based on 2D candlestick chart data. We evaluated our model for five high-demand stocks (Apple, Tesla, IBM, Amazon, and Google) and determined that our collaborative network achieved promising results and compared favorably against single-network models using either sentiment data or candlestick charts alone. The proposed method obtained the most favorable performance with 75.38% accuracy for Apple stock. We also found that the stock price prediction achieved more favorable performance over longer periods of time compared with shorter periods of time.

Short and Medium-Term Prediction of Winter Wheat NDVI Based on the DTW–LSTM Combination Method and MODIS Time Series Data

The normalized difference vegetation index (NDVI) is an important agricultural parameter that is closely correlated with crop growth. In this study, a novel method combining the dynamic time warping (DTW) model and the long short-term memory (LSTM) deep recurrent neural network model was developed to predict the short and medium-term winter wheat NDVI. LSTM is well-suited for modelling long-term dependencies, but this method may be susceptible to overfitting. In contrast, DTW possesses good predictive ability and is less susceptible to overfitting. Therefore, by utilizing the combination of these two models, the prediction error caused by overfitting is reduced, thus improving the final prediction accuracy. The combined method proposed here utilizes the historical MODIS time series data with an 8-day time resolution from 2015 to 2020. First, fast Fourier transform (FFT) is used to decompose the time series into two parts. The first part reflects the inter-annual and seasonal variation characteristics of winter wheat NDVI, and the DTW model is applied for prediction. The second part reflects the short-term change characteristics of winter wheat NDVI, and the LSTM model is applied for prediction. Next, the results from both models are combined to produce a final prediction. A case study in Hebei Province that predicts the NDVI of winter wheat at five prediction horizons in the future indicates that the DTW–LSTM model proposed here outperforms the LSTM model according to multiple evaluation indicators. The results of this study suggest that the DTW–LSTM model is highly promising for short and medium-term NDVI prediction.

Empirical greenhouse gas assessment for flexible bioenergy in interaction with the German power sector

Wind and solar power are already the major pillars of renewable power generation in Germany and will become even more dominant in the future. At the same time, dispatchable power plants phasing out. The expected increase of fluctuations in the residual load could be partial balanced by flexible bioenergy. However, there is currently no assessment approach for quantifying the systemic GHG impacts for flexible bioenergy generation.Based on the merit order concept, we develop an empirical approach to systemically assess the GHG emissions impact for electricity generated by flexible bioenergy plants. We estimate price response functions using the historical data of market prices and feed-in time series for the different forms of dispatchable non-renewbale dispatchable generation (NRDG). By calculating the expected NRDG based on these functions, and using specific emission factors, we are able to estimate the net impact for a switch from invariable to flexible bioenergy generation.The calculated net impact ranges from −20 to −36 g CO2eq per kWh, which is equivalent to a benefit of −10% to −18% respectively for an average carbon footprint of 200 g CO2eq per kWh. The calculation tools are written in Python and freely accessible on ZENODO and GitLab.

Prior informed regularization of recursively updated latent-variables-based models with missing observations

Many data-driven modeling techniques identify locally valid, linear representations of time-varying or nonlinear systems, and thus the model parameters must be adaptively updated as the operating conditions of the system vary, though the model identification typically does not consider prior knowledge. In this work, we propose a new regularized partial least squares (rPLS) algorithm that incorporates prior knowledge in the model identification and can handle missing data in the independent covariates. This latent variable (LV) based modeling technique consists of three steps. First, a LV-based model is developed on the historical time series data. In the second step, the missing observations in the new incomplete data sample are estimated. Finally, the future values of the outputs are predicted as a linear combination of estimated scores and loadings. The model is recursively updated as new data are obtained from the system. The performance of the proposed rPLS and rPLS with exogenous inputs (rPLSX) algorithms are evaluated by modeling variations in glucose concentration (GC) of people with Type 1 diabetes (T1D) in response to meals and physical activities for prediction windows up to one hour, or 12 sampling instances, into the future. The proposed rPLS family of GC prediction models are evaluated with both in-silico and clinical experiment data and compared with the performance of recursive time series and kernel-based models. The root mean squared error (RMSE) with simulated subjects in the multivariable T1D simulator where physical activity effects are incorporated in GC variations are 2.52 and 5.81 mg/dL for 30 and 60 mins ahead predictions (respectively) when information for all meals and physical activities are used, increasing to 2.70 and 6.54 mg/dL (respectively) when meals and activities occurred, but the information is withheld from the modeling algorithms. The RMSE is 10.45 and 14.48 mg/dL for clinical study with prediction horizons of 30 and 60 mins, respectively. The low RMSE values demonstrate the effectiveness of the proposed rPLS approach compared to the conventional recursive modeling algorithms.