بکارگیری نظریه مشخصات بازار برای پیش بینی شاخص بازار آینده تایوان

This research applies a market profile to establish an indicator to classify the correlation between the variation in price and value with the stock trends. The indicator and technical index are neural network architecture parameters that assist to extrapolate the market logic and knowledge rules that influence the TAIEX futures market structure via an integral assessment of physical quantities. To implement the theory of market profile on neural network architecture, this study proposes qualitative and quantitative methods to compute a market profile indicator. In addition, the indicator considers the variation and relevance between long-term and short-term trends by incorporating the long-term and short-term change in market in its calculation. An assessment of forecasting performance on different calculation approaches of market profile indicator and technical analysis is conducted to differentiate their accuracies and profitability. The experimental results show the qualitative market profile indicator outperforms the quantitative approach in a short-term forecast period. In contrast, the quantitative market profile indicator has a better trend-predicting ability, thus it is more effective in the long-term forecast period. The integration of market profile and technical analysis surpasses technical analysis as a neural network architecture parameter by effectively improving forecasting performance and profitability.

The Taiwan Futures Exchange (TAIFEX) was established in September 1997. The Taiwan Weighted Stock Index Futures (TAIEX futures) was launched in July 1998, and declared the official start of Taiwan’s futures market. A number of commodity futures electronic futures, financial futures, small TAIEX Futures were launched over a ten year period. Investors can select the appropriate investment vehicles depending on the degree of risk. Futures accounts are growing year by year. The futures market has tended to improve, becoming an important hedging and arbitrage method for Taiwan stock market investors with tools. However, the futures market has become increasingly volatile, with the legal entity participation in the futures market gradually increased (Lien, Lim, Yang, & Zhou, 2013). In addition, the financial tsunami of capital withdrawals during share transactions increasing year by year, has become the main force leading the Taiwan stock market ups and downs. Studies have shown that, in the Taiwan stock market, the average retail investor withstands losses of about 3.5% per year, with corporate investors, due to rare information and chip advantages, can obtain a 1% after-tax return (Barber, Lee, Liu, & Odean, 2004). Yu and Huarng (2008) proved fuzzy time series models can forecast TAIEX futures markets (Yu & Huarng, 2008). Steidlmayer (1984) proposed the market profile theory; refute the efficient market and random walk theory (Steidlmayer, 1984). At different time intervals, participants in different prices bid passive or active, leading to price movement rather than random development. Different participants have different thoughts and behavior preferences for the same price, so the market cannot meet the needs of each participant, without any prices representing fair value. In other words, the market is not efficient. Steidlmayer also pointed out that the risk and reward in the market is not a linear relationship (Roll, Schwartz, & Subrahmanyam, 2007). The asymmetric opportunities, irrational human investment behavior cause market fluctuations through an understanding of long-term (artificial person) and short-term (retail investors) market behavior and logic is able to predict changes in the market structure to reduce investment risk. In the face of these non-linear problems, artificial intelligence (AI) methods learn the knowledge (Lin et al., 2012 and Won et al., 2012) and rules that can be effective in predicting an environment of uncertainty without the need to rely on subjective judgment is better than the traditional model (Desai et al., 2011 and Roon et al., 2000). The neural network (NN) deals with knowledge problems as a forecasting tool (Esichaikul and Srithongnopawong, 2010 and Kaastra and Boyd, 1995). The NN through self-learning creates learning through repeated historical data (Gregoriou et al., 2007 and Ntungo and Boyd, 1998), establishing a non-linear prediction model (Yoon & Swales, 1991). The market profile concept has been widely used in the financial decision-making field (Canoles, Thompson, Irwin, & France, 1998); however, there has been little direct research (Dalton, Dalton, & Jones, 2007). Therefore, this study used the market profile principle and technical analysis (Taylor & Allen, 1992), as back-propagation neural network (BPNN) input variables. A better model than the old learning model is constructed using only the technical analysis and a new research model to explore the market logic and knowledge rules (Edwards & Magee, 1997). The market contour theory with technical analysis is extracted from the relationship between price and value using the NN knowledge of rule changes learning using the market logic and market structure (Grudnitski & Osburn, 1993). How the market profile is used as the BPNN input variables is the focus of this study. The experimental design involved observed the market profile information. The impact on the future Taiwan stock market trend, to further assess and validate the predictive ability of the different intervals, to provide an innovative investment tools for investors and future researchers as a reference.

This study used market profile theory to determine the price, value changes and trends through indicator values establishment as a basis for the relationship between the technical specifications for the aggregate evaluation of the physical forces. Different market profiles were compared and calculated with the simple use of the differences between the technical analysis models. The experimental results are at the predicted time of 5 and 15 min. Qualitative computing market profile deviation value, the accuracy rate is better than quantitative way. At the predicted time of 30 min, calculated using quantitative method superior qualitative. The main cause of the qualitative methods can learn effectively break through and below the status of market pressure and support, but due to the qualitative calculation, take the firstorder and second variable, cannot reflect the true market behavior and therefore judged that the long-term trend decreased ability. The quantitative method due to short-term correction in the case of pressure support caused by short-term forecasts cannot effectively enhance interpretation of ability but the trend is better than qualitative methods, so the long-term prediction is significantly improved. Whether using qualitative or quantitative calculation, by adding a short-term 15-min market profile indicators help to improve the forecasting performance, especially with five minutes to predict the time of the most significant enhancement effects. Due to short-term 15-min market profile indicators for predicting stock prices of longer lines, judged weak. Therefore, when the forecast longer hours, the relatively small effect, and even negative growth. Comparing the technical indicator with the neural network input variables, the experimental results confirm that adding the market profile indicators can effectively improve the forecast accuracy and profitability. To predict the change range after 30 min to enhance the most significant effect. From experiments designed to observe the market profile information, and the future trends of the Taiwan stock market, to further assess and validate the predictive ability of the different intervals, to provide an innovative investment tools for investors and future researchers. Experimental results show qualitative market profile indicator outperforms quantitative approach in short-term forecast period. In contrast, quantitative market profile indicator has a better trend-predicting ability thus it is more effective in long-term forecast period. The results also manifest that both approaches considering the combination of long-term and short-term change in market enhance forecasting performance and are most effective in short-term time interval. In conclusion, the integration of market profile and technical analysis surpasses technical analysis as a parameter to neural network architecture by effectively improving forecasting performance and profitability.