Forecasting the Highest and Lowest Prices in Financial Markets Using a VMD-LSTM Hybrid Model

An effective financial market trading decision is usually dependent on superior forecasting. Forex market as the largest financial market is chosen in this study. The main objective of this paper is to explore the forecasting performance of the proposed multiple-price model which integrates close, low and high price information, by using Artificial Neural Network (ANN). The architecture of the network and the related algorithms are described. The effects due to different choices of preprocessing methods, combinations of input variables and different time intervals of forecasting are examined. By using the best multiple-price model, trading strategies with high and low prices are developed as well. The results have shown that in terms of both absolute values and trends of the prices, forecasting accuracy has improved compared with single price model. This is especially so for low and high prices whose directional accuracies are much higher. The trading performance is also proven to have much better total return than buy & hold strategy, and trading with high price has the best overall performance considering both return and risk.

Sellers are typically better informed about product quality than their customers. Because sellers have an incentive to misrepresent quality, it may not be possible for market prices to effectively convey this information to rational consumers, as was first argued by Akerloff (1970). The purpose of this paper is to argue that even if sellers are initially better informed than buyers, prices may yet be informative if buyers can purchase additional information about quality from an external, reliable source. In this setting, the informative role of prices is shown to depend crucially on the cost of external information to consumers. In particular, there exists a critical value such that when the cost of information is below this value, the market equilibrium is characterized by two distinct prices and a different pricing strategy for each type of seller. High quality sellers deterministically charge the high price while low quality sellers randomize between the low price, which corresponds to the low quality price that obtains under conditions of complete information, and the high price. The equilibrium frequency with which the high price is mimiced by low quality sellers decreases as the cost of information grows smaller and goes to zero in the limit. Correspondingly, the level of the high price increases as the cost of information decreases and approaches the complete information high quality price in the limit. Thus the less costly it is for buyers to become independently informed, the less noise the low quality seller generates and the more informative about quality the high price is.

Bitcoin has attracted the attention of investors lately due to its significant market capitalization and high volatility. This work considers the modeling and forecasting of daily high and low Bitcoin prices using a fractionally cointegrated vector autoregressive (FCVAR) model. As a flexible framework, FCVAR is able to account for two fundamental patterns of high and low financial prices: their cointegrating relationship and the long memory of their difference (i.e., the range), which is a measure of realized volatility. The analysis comprises the period from January 2012 to February 2018. Empirical findings indicate a significant cointegration relationship between daily high and low Bitcoin prices, which are integrated on an order close to the unity, and the evidence of long memory for the range. Results also indicate that high and low Bitcoin prices are predictable, and the fractionally cointegrated approach appears as a potential forecasting tool forcryptocurrencies market practitioners.

Despite improvements in sanitation and water supply, cholera remains a serious public health burden. Vaccination is included among recommendations for cholera control. Cultural concepts of illness are likely to affect vaccine acceptance. This study examined social and cultural determinants of anticipated acceptance of an oral cholera vaccine (OCV) prior to a mass vaccination campaign in Zanzibar. Using a cultural epidemiological approach, 356 unaffected adult residents were studied with vignette-based semi-structured interviews. Anticipated acceptance was high for a free OCV (94%), but declined with increasing price. Logistic regression models examined social and cultural determinants of anticipated acceptance at low (USD 0.9), medium (USD 4.5) and high (USD 9) price. Models including somatic symptoms (low and high price), social impact (low and medium) and perceived causes (medium and high) explained anticipated OCV acceptance better than models containing only socio-demographic characteristics. Identifying thirst with cholera was positively associated with anticipated acceptance of the low-priced OCV, but acknowledging the value of home-based rehydration was negatively associated. Concern about spreading the infection to others was positively associated at low price among rural respondents. Confidence in the health system response to cholera outbreaks was negatively associated at medium price among peri-urban respondents. Identifying witchcraft as cause of cholera was negatively associated at medium and high price. Anticipated acceptance of free OCVs is nearly universal in cholera-endemic areas of Zanzibar; pre-intervention assessments of community demand for OCV should not only consider the social epidemiology, but also examine local socio-cultural features of cholera-like illness that explain vaccine acceptance.

Stock market price prediction is vital for investment decision amid difficulties with effective price predictions. The paper aims to analyse the rate of effectiveness in actual stock market price prediction using the open, high and low prices. The paper draws insight from diverse prior research with assorted models such as Markov Chain, time series and computer aided stock price prediction. The paper’s approach is quantitative with forty-three days stock market price data from S&P500 and Shanghai Composite Index. Data was analysed with the regression statistics. Results show that the open, high and low prices can significantly predict the actual market price at probability level of P<0.0001 for both the S&P500 Index and the Shanghai Composite Index. Prediction rates exceed 70% for S&P500 and over 80% for Shanghai Composite Index. The model was verified by using data other observation periods (during the COVID-19 and during the financial crisis). The implication therefore is that in the absence of other expensive market information, an average investor may use the open, high and low prices to make a useful prediction of actual stock market price. The findings present a useful case reading for academics in business schools and offer an agender for future research to apply this model in other stock markets. The paper offers a novel value from the finding by demonstrating that the showing that application of open, high and low prices with regression may give a prediction accuracy rate of over eighty percent, which is higher than reported seventy percent prediction rate in prior work that used other models.

Forex inaugurates new plant

FLF-LSTM: A novel prediction system using Forex Loss Function

IA N ILLUSTRATION of Common Stock Price Changes. The kind of price _ changes which will be discussed in this paper are presented in Figure 1. In this figure we have graphs of the annual high and low prices of two common stocks, Inland Steel, a high priced issue, and Interlake Iron, a relatively low priced stock. At the bottom of Figure 1 the actual prices are plotted and from this one can see that Interlake Iron fluctuates less in absolute amount than Inland Steel. In the middle of Figure 1 the price level is taken into account by plotting the logarithms of the annual high and low prices. Thus, the spread between the two lines measures percentage changes and the graph reveals that such changes in price tend to be larger for the low priced stock. At the top of Figure 1 the square roots of the annual high and low prices are plotted and the resulting picture suggests that the fluctuations in the square root of prices are independent of price level. According to the above discussion of Figure 1, price level seems to be a factor in determining price change in common stocks. The question that arises is whether or not there is a consistent relationship between price change and price level. Purpose of this Study. The general purpose of this study is to investigate the likelihood of a relationship between price change and price level of common stocks. The problem, as it is treated in this investigation, consists essentially of three parts: (1) Measuring relationships by means of statistical analysis; (2) Investigating the significance of such relationships; (3) Setting forth some implications of such relationships. Summary of Related Studies. The problem of studying the relationship of price change and price level of common stock prices is related to and was suggested by articles which discuss the tendency of per cent changes in low priced stocks to be higher than the per cent changes in high priced stocks. Along this line some writers have pointed out that changes in the square roots of prices tend to be constant. D. W. Ells-

This paper addresses the modeling and forecasting of daily high and low asset prices in the Brazilian stock market using a fractionally cointegrated vector autoregressive model (FCVAR). Forecasts are then used in a simple trading strategy to evaluate the application of technical analysis (TA) for equity shares traded at the B3. As a flexible framework, FCVAR is able to account for two fundamental patterns of high and low asset prices: their cointegrating relationship and the long-memory of their difference (i.e., the range), a measure of realized volatility. The analysis comprises the twenty most traded stocks at the B3 during the period from January 2010 to May 2017. Empirical findings indicate a significant cointegration relationship between daily high and low prices, which are integrated of an order close to the unity, as well as the range displays long memory and is in the stationary region in most of the cases. Based on historical data, results support that the high and low prices of equity shares are largely predictable and their forecasts can improve TA trading strategies applied on Brazilian equity shares. Further, the fractionally cointegrated approach appears as a potential forecasting tool for market practitioners on their investment strategies.

Day-ahead electricity price forecasting (DAEPF) is vital for participants in energy markets, particularly in regions with high integration of renewable energy sources (RESs), where price volatility poses significant challenges. The accurate forecasting of high and low electricity prices is particularly essential, as market participants seek to optimize their strategies by selling electricity when prices are high and purchasing when prices are low to maximize profits and minimize costs. In Japan, the increasing integration of RES has caused day-ahead electricity prices to frequently fall to almost zero JPY/kWh during periods of high RES output, creating significant profitability challenges for electricity retailers. This paper introduces novel custom loss functions and metrics specifically designed to improve the forecasting accuracy of extreme prices (high and low prices) in DAEPF, with a focus on the Japanese wholesale electricity market, addressing the unique challenges posed by the volatility of RES. To implement this, we integrate these custom loss functions into a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model, augmented by an ensemble learning approach and multimodal features. The proposed custom loss functions and metrics were rigorously validated, demonstrating their effectiveness in accurately predicting high and low electricity prices, thereby indicating their practical application in enhancing the economic strategies of market participants.

The prediction of futures prices is a great challenge. On the other hand, it can bring investors great profits. Most researches just show the predictions of closing prices but we can also predict high and low prices. The high and low prices have lower noises than closing prices, making it easier to predict them and to use them for making profitable strategies. In this paper, we build a model to predict high and low prices of soybean futures with the LSTM neural network using the dataset from the Dalian Commodity Exchange. Then we use mean absolute error (MAE) and trend accuracy to evaluate the performance of this model. For comparison, we predict the closing price using the LSTM neural network and build another prediction model based on the BP neural network. Results show that we get higher accuracy predicting the trends of high and low prices. Also, the prediction model based on the LSTM neural network performs better and it gets more than 80% of the accuracy in trend estimation when the predicting high prices or low prices have high volatilities.

An essential aspect of stock trading is the accurate forecast of stock price.This enables buy and sell points to be determined, which facilities profitability whilst reducing potential loses.This paper proposes a "Two-stage pattern Strategy (TSPS)" as an effective and intuitive mechanism to identify buy and sell signals.Utilizing technical analysis methods and pattern recognition principles, the TSPS indicates that (1) the trend to be more important than the isolated price; (2) two continuous unidirectional trends could verify the uptrend or downtrend; (3) the high or low price have more prediction power than the closing price; (4) the low price is more effective in prediction in the uptrend case, while the high price is more valuable than the low price in the downtrend case.Accordingly, this paper establishes mechanism to recognize up or down patterns and the pinpoint for buying or selling.This empirical study was done to verify the prediction power and profitability of TSPS.This study compared the performance of the U.S. and Chinese markets.The results show that TSPS can be widely used in the market regardless of the economic environment

Social and cultural features of vaccine acceptance and cost-effectiveness of an oral cholera mass vaccination campaign in Zanzibar

This chapter develops a dynamic systems analysis to the financial system that generates financial market clearing prices over time and circumstances. It also develops a constructive critique that addresses the theoretical roots of fair value in the efficient financial market hypothesis. The fair value basis of accounting abstracts away from realization. Accounting information provision does result in playing a better role when it provides a fundamental signal that remains independent from financial market price dynamics. Equilibrium does assume the existence of efficient market prices, scoping out all the details concerning their formation through ignorance and hazard. The market prices are assumed to exist and to comply with conditions derived from equilibrium in the financial market. The financial market price formation process involves interactions and exchanges between investors, who aim to buy and selling securities traded on the financial market of reference. The efficient financial market hypothesis develops an elegant mathematical model that exploits some properties of equilibrium prices.

Objectives: The proposed work integrates multiple Machine Learning approaches in a single model to be used for the analysis and forecasting of Time Series data. Methods: In the present work, the concept of Stacked Ensemble learning is proposed that uses Multiple Linear Regression and Support Vector Regression techniques as the base models. A Meta Model is constructed based on Multiple Linear Regression with necessary modifications. The outputs from the base models are fed into the meta-model which is mended with the capability of combining the predictions from the two base models to produce better results than the individual constituent parts, after running a k-fold training procedure. Findings: The proposed model is capable of analyzing and predicting any Time Series data. In the present study, stock data of six companies enlisted in the National Stock Exchange of India are analyzed for the prediction of the next day’s Open, High, and Low prices. The proposed work achieves better accuracy and reduces the error in prediction when compared to similar works done in the same field. Novelty: The amalgamated technique used in this work can be considered as a generalization of the stacked ensemble method in a broader aspect. The proposed model combines the strengths of multiple Machine Learning methods into a single model to achieve better performance than its individual counterparts. Further, several recent works have tried to predict only the next day’s Open and Closing Prices of stocks, but for an intraday trader, prediction of the next day’s Low and High prices of a stock are more significant than the closing prices. Very few works have predicted all of the Open, High and Low prices in a single study, our present work achieved this quite successfully. Keywords: Machine Learning; Stacked Ensemble Model; Support Vector Regression; Multiple Linear Regression; Time Series data analysis; Stock Price Prediction