Modification of technical analysis indicators and increasing the rate of return on investment

Identification of factors determining the level of prices in financial markets is of interest to many researchers. Investors, in order to increase their profits from investments, expect tools and techniques that allow them to effectively and reliably predict price changes. Numerous polemics in this regard are available in the literature on the subject (Reilly & Brown, 2011). Two analyses are used to forecast prices of stocks: fundamental and technical. Despite the many differences in these approaches, the supporters of each of these analyses agree on the key factors explaining the level and trends of prices in financial markets. The relationship between stock prices and factors in fundamental analysis is described, among others, by Gordon and Shapiro (1956). In the field of fundamental analysis, one of the researchers who provides the most significant research is J.A. Ohlson (1995). He formulated a model that indicated the share price as a key factor determining an investor’s profit. Using Ohlson’s model is a recognised method of stock valuation. The relationship between profit and share price is also indicated by O’Collins (1994, 2011). As part of fundamental analysis, two approaches are distinguished in the literature (Figure 1).

Figure 1.

Fundamental analysis is an analysis that is based on many economic measures and indicators (Jajuga, 2015; Małachowski & Gadowska-dos Santos, 2021). The presented approaches (Figure 1; top-down approach) indicate that the investor (or potential investor) starts with an analysis of economic conditions, then an analysis of the economic environment. The final stage of this approach is the organizational analysis of the company in question. This approach is used in the event that an investor is looking for the best stocks in which to invest. In the bottom-up approach, the opposite analysis occurs—starting with company analysis.

An alternative to fundamental analysis is technical analysis. It is defined as the study of any market by using historical information about prices and trading volumes to predict future price changes (Moghaddam & Momtazi, 2021). Murphy (2017) defines this method of price forecasting as information about prices, number of transactions, and existing transactions in order to predict the price trend in the future. The advantage of this method is that the trader can use technical analysis to predict the direction of price movements, thus achieving an effective decision (Elbialy, 2019). Price prediction using technical analysis may consist of (1) trend analysis or (2) analysis of indicators (Figure 2).

Figure 2.

Technical analysis is mainly used to forecast changes in the prices of selected financial assets by investors investing their money in various instruments, most often on the stock exchange or the currency market (Dolan, 2011; Lim, 2015; Rockefeller, 2019). As part of technical analysis, various indicators and candle formations are used, and support (or resistance) zones are determined. Technical analysis indicators used to forecast prices on financial markets can be divided into several categories. The first of them are oscillators, i.e. indicators that oscillate between certain values that are used to read discrepancies, neutral market conditions, or estimates (Shannon, 2008; Kahn, 2011). The most popular oscillator indicators include the RSI and the Stochastic Oscillator (STS). In the case of price forecasting more volatile assets, momentum indicators can be used which show the dynamics of price change, placing more weight on price amplitude and the periods of analysis closest to the current one (Dunham, 2012). Popular indicators in this category are the momentum index and CCI (Commodity Channel Index).. In order to determine what price trends will prevail in a given market, you can use indicators typically provided for determining the trend in which a given market is located. Among the most popular in this group, Bollinger Bands, or, for example, Parabolic SAR, can be distinguished (Borowski, 2017). Please note that the indicators indicate only potential, projected price changes. Technical analysis is based on the assumption that prices or exchange rates are subject to trends, and its assumptions are to help determine whether the market trend will continue when it reverses or a completely new one will be established. The task of technical analysis indicators is therefore to, among other things, help determine the moment of trend reversal. Technical analysis focuses the investor’s attention on price changes, not on fundamental aspects of the valuation of a given company or asset, as is the case with fundamental analysis (Murphy, 2017). It should be emphasised that there is no indicator characterised by flawlessly generated trading signals (Plummer, 2010). There are many technical analysis indicators, and each of them can be calculated on the basis of individually selected parameters, which in practice implies almost unlimited possibilities of personalisation of the indicator in order to best match the analysed asset (Grimes, 2012; Hassen, 2017). However, it should be remembered that even if the historical analysis of a selected asset clearly indicates a given investment strategy as potentially profitable, the market may function completely differently in the future. This means that assumptions based on historical price movements may result in a loss. Summing up the theoretical considerations, it can be said that in the literature on the subject, fundamental analysis and technical analysis are not treated as complementary techniques. Researchers suggest conducting two analyses in parallel to achieve the optimal effect of investment plans (Drakopoulou, 2016). The second solution leading to the optimisation of investment profits is a two-stage operation: (1) fundamental analysis (price modelling) and (2) analysis of price changes—technical analysis (Schwager, 1984). Such a hybrid approach for investors would be an optimal solution. However, it should be remembered that not every investor (primarily individuals) has the ability to carry out modelling, the use of fundamental analysis. Therefore, technical analysis provides more opportunities for such investors.

Technical analysis indicators are used to forecast trends and future price changes also on the Warsaw Stock Exchange. The Polish market is run by the Polish Securities Depository (GPW) and the National Depository for Securities in Warsaw (GDPW). GPW organises stock exchange trading in financial instruments, and GDPW is a capital market infrastructure institution operating in the area of securities registration, transaction settlement and services related to transaction reporting. The above authorities are supervised by the Polish Financial Supervision Authority (KNF), which also performs supervisory tasks delegated by the European Securities and Markets Authority (ESMA) (Banaszczak-Soroka, 2019). The over-the-counter market (OTC), on the other hand, is an alternative exchange system, which means that it is not subject to such restrictive requirements, although it is regulated by relevant regulations and authorities—a.o. restrictions on leverage or a complete ban on offering certain instruments (e.g. binary options). The most popular over-the-counter market is Forex. Trading on this market takes place over-the-counter, as transactions are carried out outside a centralised exchange, and this market does not have a physical headquarters. The Forex market operates in the mode “around the clock,” which means that transactions can be conducted 24 hours a day; thanks to that, it is often used for intraday trading (Noonan, 2022). A very important role in the popularisation of the OTC market is played by a wide range of available instruments, as well as relatively low transaction costs, amounting even to less than 0.01% of the contract value (Cuthbertson, Nitzsche, & O’Sullivan, 2019). Investors can buy and sell currencies through a network that directly connects individual banks, traders, and brokers. In the OTC market, buying and selling transactions are concluded on the basis of bilateral agreements between participants who set their terms without the intervention of a centralised regulatory authority. Today’s market of OTC financial derivatives allows trading in many different types of complex derivatives without the need to obtain specialised certificates, authorisations, or permits. This is one of the aspects affecting its growing popularity (Czekaj, 2017). This market is characterised by the highest degree of internationalisation among all segments of the world’s financial markets. When comparing the regulated market, offering the underlying instruments to the OTC market, the concentration of capital is much higher in the case of the derivatives market (Mishkin & Eakins, 2021). It should be noted, however, that all securities and derivatives related to the financial turmoil, which began with the collapse of the U.S. mortgage market in 2007, were traded on OTC markets (Stępień & Kawa, 2015). Trading in OTC instruments is enabled through more than 40 investment banks, and most of the trading is generated by a group of 16 of the largest providers (G-16 liquidity dealers). These entities are classified as systemically important (G-SIBs), due to their majority share in the global derivatives market. This group consists of well-known transnational corporations, such as: Bank of America, BNP Paribas, Barclays, Citigroup, Credit Suisse, Goldman Sachs, JP Morgan Chase, HSBC, Morgan Stanley, Merrill Lynch or UBS, and Wells Fargo (Fałat-Kilijańska et al., 2017; Hull, 2023).

In order to minimise the impact of emotions and relieve the investor, who is often somehow attached to an open position, the aforementioned Take Profit and Stop Loss orders were created For investment strategies that have a predetermined level of loss and profit, the trader can activate automatic orders that will be executed when the instrument reaches a certain price. Take Profit (TP) orders are definitely more optimistic because they are related to profit taking. The possibility of using the TP order allows investors to automate trading on the market. A Forex trader, after receiving a signal to take a long position based on his investment strategy, sets an automatic TP order at a given level and can wait for events to develop by performing other actions. If the exchange rate reaches the indicated level, the position will be automatically closed, and the investor will make a profit—this is how the TP order works. Similarly, Stop Loss (SL) orders are not responsible for taking profits and losses. These orders are intended to close a position at a predetermined loss. Returning to the previously mentioned example of an investor and his strategy based on the EUR/USD currency pair, let us assume that the strategy, apart from the optimistic variant—profit taking, also provides for a pessimistic variant—is cutting losses. In this case, the investor’s task is to close the losing position at the moment of negative deviation of the price by a given value from the price from the moment of the trading signal. In order to not observe the chart or stress unnecessarily, the investor, using the option of setting a SL order, activates its level at the price determined in accordance with his strategy. If the price falls to this level, the position will be automatically closed and the trader will incur a loss—this is how SL orders work (Surdel, 2006; Mladjenovic, Brooks & Dolan, 2021).

The data used to conduct the research came from the publicly available resource of the MetaTrader4 platform from the period of September 1, 2021 to February 10, 2023. The above-mentioned data source was used due to easy access and relatively high prevalence of the mentioned platform among investors.

The subject of the research was indications of the investment strategy based on technical analysis indicators, Parabolic SAR, and RSI in the context of their accuracy, measured on the basis of correctness of indicating the future change in the EUR/USD currency pair.

The analyses were carried out over a period of 4 months (September–December 2022) for the first and second strategies, and a period of 17 months (September 2021–February 2023) for the third strategy. The discrepancy in the duration of the analysis period is related to the frequency of generated trading signals. As verified during the conducted analyses, the first and second strategies generated almost 1.5 transactions per day in the period under review, which is more than 4 times more than the third strategy which generated 0.34 signals per day. Due to this discrepancy and with a view to standardising the number of analysed signals, in order to make the best possible comparison of the above-mentioned strategies, the observation period for the third strategy was extended accordingly to obtain an equal number of trading signals.

To build an exemplary strategy, technical analysis indicators, i.e. Parabolic SAR and RSI, were used. The strategy was developed for use in the Forex market, where the Euro-Dollar (EUR/USD) currency pair was used to test it. The strategy was applied on a 1-hour time series (H1). Three modifications of this strategy were used. The results achieved by the above-mentioned strategies in the period of September–December 2022 in the case strategies one and two and the period from September 2021 to February 2023 in the case of the third strategy were compared. The comparison of parameters and assumptions of the analysed strategies is as follows:

Strategy 1: Basic strategy based on the default setting of Parabolic SAR and RSI. The position is opened in accordance with the default settings for the above-mentioned indicators, i.e. a)

Long position: entering into a trade at the time of closing the H1 candle, at the same time with the RSI indicator level above 50 points and the position of Parabolic SAR below the price chart. The transaction is closed when the position of the Parabolic SAR indicator changes above the price chart—an indication to take a short position.

Closing of a transaction in the third strategy: a)

Long position: taking a long position occurs when the Parabolic SAR indicator is placed below the price chart, with the RSI indicator level greater than or equal to 60 points.

Closing a transaction can occur in several cases:

An unfavourable price movement after the conclusion of the transaction (increase in the exchange rate in the case of a short position or decrease in the case of a long position), and the related change in the value of the RSI indicator at the close of a given H1 candlestick, at the level of more than 50 points (short position) or at a level below 50 points (long position), causes the closing of the position at the current level of loss, together with the closing of the first H1 candle, at which the RSI exceeds the given level.

On the basis of the above assumptions, analyses were made of the third strategy during the period considered.

This study uses data characterising changes in the EUR/USD exchange rate. These were the opening price and the closing price on a given day of quotations of the EUR/USD currency pair or the average price, respectively for a given period. In addition, the analysis concerned (%) changes in the EUR/USD exchange rate. An appreciation of the Euro was observed in the case of an increase in the value of the exchange rate and depreciation in the event of its decrease. Volume determines the average of the numbers of transactions concluded on this currency pair over a given period. The average value of the opening and closing prices, the change (%) directly related to the EUR/USD exchange rate. The number of transactions concluded on the EUR/USD currency pair are presented in Table 1.

In 2021 and 2022, we observed a decrease in the value of the EUR/USD currency pair. At the beginning of the period, in September 2021, the closing price was 1.1764, and in February 2023 the closing price fell to 1.0783. It is worth noting that in the analysed period the downward trend was not homogeneous, and the exchange rate showed some volatility. In September and October 2022, there was a situation of parity on the above-mentioned currency pair, i.e. 1 euro was equal to 1 dollar. Percentage changes for individual months are relatively small and range from −0.21% to 0.24%. These changes indicate a small monthly volatility of the EUR/USD currency pair. The average trading volume of the EUR/USD currency pair also shows some volatility during the period under review. The lowest average volume occurred in October 2021 (62.28) and the highest in October 2022, reaching a value of 92.18. It is worth noting that there is no clear correlation between the percentage change and the volume of trade.

To summarise, the analysis of data from the table on the EUR/USD currency pair indicates a downward trend in the analysed period. Monthly percentage changes are relatively small, indicating moderate volatility of the currency pair. The trading volume also shows volatility, but there is no clear relationship between the percentage change and the trading volume. The mean and median of all described variables (in the entire analysed end) of both closing and opening are similar, which suggests that there are no large differences between the values (Table 2).

The determined coefficients (skewness and kurtosis) for the variables accepted for analysis indicate asymmetry of the distribution. For closing and opening prices, the skewness value is negative, which means that the distribution is left-sided. A low kurtosis (−1.07920) indicates that the values are scattered over a wide range. The average value of the percentage change (Euro appreciation/depreciation) indicates that there are no significant overall upward or downward trends in the analysed dataset. For a change in the EUR/USD exchange rate, the value of skewness (0.305433) is positive, indicating a right-hand asymmetry in the distribution. Kurtosis (0.87710) is relatively low, which means that there are more extreme outliers in the analysed dataset than in the normal distribution. The average and median for the number of transactions concluded on the EUR/USD currency pair are similar, so there are no large differences between the volume values. The skewness for this parameter (0.377485) indicates a right-sided asymmetry of the u distribution, while kurtosis (1.630956) is higher than for other variables, so in the volume distribution most of the values are concentrated around the mean.

The research was conducted on the MetaTrader4 (MT4) platform, which is one of the most popular online trading platforms. It offers the possibility of trading using leverage, mainly in the Forex currency market and the Contracts for Difference (CFD) market. All transactions made through the MetaTrader platforms are made online. The analysis covered a period of 4 months (September–December 2022) in the case of the first and second strategies, and a period of 17 months (September 2021–February 2023) in the case of the third strategy, where the observation time was extended due to the lower average number of generated transaction signals. The results of individual transactions achieved under individual strategies are presented in Figures 3–5.

Figure 3.

Figure 4.

Figure 5.

Analysing the data presented in Figures 3 and 4, one can see similarities in frequency and values in the case of profitable trading signals. Due to the SL level, nine potentially profitable trades were closed when they turned negative. A significant difference, at first glance, can be seen on the losing (negative) side of transactions. The rigidly set SL meant that relatively numerous transactions below −200 points occurring in the first strategy did not take place, and therefore there were no excessive losses in the second strategy. As can be seen from the analysis of the aggregate results presented in Table 3, the SL level applied significantly contributed to achieving a favourable outcome when using the second strategy compared to the first one.

As in the case of the first two strategies, analysing Figure 5 relating to the third strategy, one can see a dominant number of losing trades (72%), while correspondingly higher profitable transactions made the overall result of the strategy unambiguously favourable. Figure 3–5 show the results achieved by the analysed strategies 1, 2, and 3 broken down into individual transactions. While in the case of the first two strategies, the occurrence of strong upward or downward trends is relatively less important, the third strategy has been developed in accordance with the slogan that the trend is the investor’s greatest “ally”. A higher number of profitable third strategy trades can be seen during periods of increased price amplitude shown in Figure 6. This is due to the fact that the TP level provided for the third strategy is relatively high, approximately 1.3–1.5% of the EUR/USD exchange rate.

Figure 6.

The daily price range of the EUR/USD currency pair, measured by the difference between the maximum and minimum prices for each day, is shown in Figure 6.

Figure 7 presents the development of the total result of the analysed strategies along with subsequent transactions.

Figure 7.

The result is presented in points, which in the case of Forex and currency pairs quoted to five decimal places, are as follows: 1000 points = 100 pips = e.g. 1 gr / 1 cent, depending on the currency pair. The parameters for the analysed strategies are presented in Table 3.

It is worth noting that despite a lower share of profitable transactions, the second strategy turned out to be much more profitable than the first strategy. The explanation of this situation is undoubtedly related to the average value of the losing and profitable positions. While the average winning position in strategy two is slightly larger than in strategy one, the almost twice lower, average value of a losing position explains such a large difference in the aggregate result of these almost “twin” strategies. Slightly different statistics were obtained after the analysis of the results of the third strategy, the individual transaction signals of which were set for a much longer execution time, as evidenced by almost 5 times less generated signals per trading day. The average value of the profitable position was also much higher. This is a deliberate effect associated with the change of indicator settings. On average, a losing position accounted for 58% of the profitable position for the first strategy, 34% for the second strategy, and 33% for the third strategy. It can be assumed that in addition to the share of profitable signals, it is the ratio of a losing to a profitable position that is crucial in the context of the profitability of a given strategy.