Performance Evaluation of Machine Learning Algorithms for Stock Price and Stock Index Movement Prediction Using Trend Deterministic Data Prediction

This experimental study addresses the problem of predicting the direction of stocks and the movement of stock price indices for three major stocks and stock indices. The proposed approach for processing input data involves the computation of ten technical indicators using stock trading data. The dataset used for the evaluation of all the prediction models consists of 11 years of historical data from January 2007 to December 2017. The study comprises four prediction models which are Long Short-Term Memory, XGBoost, Support Vector Machine ( and Random forests. Accuracy scores and F1 scores for each of the prediction models have been evaluated using this input approach. Experimental results reveal that a continuous data approach using ten technical indicators gives the best performance in the case of the Random Forest classifier model with the highest accuracy of 84.89% (average wise 83.74%) and highest F1 score of 89.33% (average wise 83.74%). The experiments also give us an insight into why a Naïve Bayes Classification model is not a suitable prediction model for the above task.

Effects of Attentional Withdrawal on Chinese Disabled Biathletes’ Shooting Performance

Under two situations of static empty gun preview and simulated competition, the validity of self-suggestion language “preloading compaction,” which suggests attentional withdrawal, was verified, respectively. Methods. The experiments in the two parts were designed in a single factor group. The participants (Chinese disabled biathletes) fired 20 times (4 groups in total, 5 times in each group) both in pretest and posttest, and they used self-suggestion before each group of firing in posttest. Optoelectronic shooting device based on wireless laser transceiver technology was selected to collect the shooting technical indicators, scores, and other information. The operation evaluation form reflected the subjective feelings of the participants, and the heart rate was measured by the chest band. Results. By the method of big data analysis, under the condition of static empty gun preview, the results showed that there was an extremely significant difference between shooting score and quality of triggering. All the marks on the operation evaluation form improved after the intervention. In the mock competition situation, shooting score has highly significant differences with quality of triggering and relative triggering value and significant difference with aiming ability. The marks of the operation evaluation form have been improved; the heart rate of firing all decreased in different degrees. Conclusions. The attentional withdrawal has positive effects on the shooting scores, technical indicators of optoelectronic shooting device, subjective feelings, and heart rate of disabled biathletes under the two conditions. TIRE (the timing of firing) is the most sensitive indicator and can be used as the focus of future training and scientific research.

Technical indicators of modern trawlers

The article identifies the problem of the lack of up-to-date publicly available statistical information on the technical and economic performance of modern fishing vessels, including foreign mining vessels, which entails serious restrictions on the choice of a prototype, from which domestic authors are forced to use outdated indicators of ship designs built in the second half of the XX century. The methods of obtaining the necessary information using the open databases of the Global Fishing Watch organization are presented. The methods of the organization's work with information and the characteristics of some databases are described. The data has been processed for further use. According to one of the fishing criteria, the most effective trawlers of 2020 were selected. Thanks to the obtained statistics on the operation of trawlers, it was possible to obtain many technical indicators of the vessels ' operation: the operating mode, the structure and duration of the fishing voyage, the form of fishing organization, the balance of calendar time. As the statistics are processed, the results are analyzed. The observations made are described: about the round-the-clock operation mode; about the duration of storming; about the associated fishing. The obtained indicators were compared with the indicators of trawler factories in 1969. The necessity of continuing the study of the main technical and economic indicators of modern fishing vessels: income, expenses, profit is justified.

Penalized Logistic Regressions with Technical Indicators Predict Up and Down Trends

Predicting up and down trends for stock prices is an important puzzle in the financial field. Hu & Jiang (2021) proposed logistic regression with 6 technical indicators to predict up and down trends for Google's stock prices. In this paper we further propose the five penalized logistic regressions with 19 technical indicators: ridge (L2), lasso (L1), elastic net(EN), smoothly clipped absolute deviation (SCAD) and minimax concave penalty (MCP) to improve the prediction accuracy. Firstly, we combine the iterative weighted least square algorithm with the coordinate descent algorithm, and apply a training set to obtain parameter estimators and probability estimators. Then we adopt a test set to construct confusion matrices and receiver operating characteristic (ROC) curves, and apply them to assess their prediction performances. Finally we compare the proposed five prediction methods with logistic regression, support vector machine (SVM) and artificial neural network (ANN) , and found that the MCP penalized logistic regression performs the best. Therefore, we develop a new efficient prediction method to predict up and down trends for stock prices.

Demand forecasting: an alternative approach based on technical indicator Pbands

Research background: Demand forecasting helps companies to anticipate purchases and plan the delivery or production. In order to face this complex problem, many statistical methods, artificial intelligence-based methods, and hybrid methods are currently being developed. However, all these methods have similar problematic issues, including the complexity, long computing time, and the need for high computing performance of the IT infrastructure. Purpose of the article: This study aims to verify and evaluate the possibility of using Google Trends data for poetry book demand forecasting and compare the results of the application of the statistical methods, neural networks, and a hybrid model versus the alternative possibility of using technical analysis methods to achieve immediate and accessible forecasting. Specifically, it aims to verify the possibility of immediate demand forecasting based on an alternative approach using Pbands technical indicator for poetry books in the European Quartet countries. Methods: The study performs the demand forecasting based on the technical analysis of the Google Trends data search in case of the keyword poetry in the European Quartet countries by several statistical methods, including the commonly used ETS statistical methods, ARIMA method, ARFIMA method, BATS method based on the combination of the Cox-Box transformation model and ARMA, artificial neural networks, the Theta model, a hybrid model, and an alternative approach of forecasting using Pbands indicator. The study uses MAPE and RMSE approaches to measure the accuracy. Findings & value added: Although most currently available demand prediction models are either slow or complex, the entrepreneurial practice requires fast, simple, and accurate ones. The study results show that the alternative Pbands approach is easily applicable and can predict short-term demand changes. Due to its simplicity, the Pbands method is suitable and convenient to monitor short-term data describing the demand. Demand prediction methods based on technical indicators represent a new approach for demand forecasting. The application of these technical indicators could be a further forecasting models research direction. The future of theoretical research in forecasting should be devoted mainly to simplifying and speeding up. Creating an automated model based on primary data parameters and easily interpretable results is a challenge for further research.

Two-Stage Hybrid Machine Learning Model for High-Frequency Intraday Bitcoin Price Prediction Based on Technical Indicators, Variational Mode Decomposition, and Support Vector Regression

Due to the inherent chaotic and fractal dynamics in the price series of Bitcoin, this paper proposes a two-stage Bitcoin price prediction model by combining the advantage of variational mode decomposition (VMD) and technical analysis. VMD eliminates the noise signals and stochastic volatility in the price data by decomposing the data into variational mode functions, while technical analysis uses statistical trends obtained from past trading activity and price changes to construct technical indicators. The support vector regression (SVR) accepts input from a hybrid of technical indicators (TI) and reconstructed variational mode functions (rVMF). The model is trained, validated, and tested in a period characterized by unprecedented economic turmoil due to the COVID-19 pandemic, allowing the evaluation of the model in the presence of the pandemic. The constructed hybrid model outperforms the single SVR model that uses only TI and rVMF as features. The ability to predict a minute intraday Bitcoin price has a huge propensity to reduce investors’ exposure to risk and provides better assurances of annualized returns.