Optimization and assessment of Trading Strategies in Russian FX and Interest Rate market

On average, foreign exchange traders viewed success through fundamental factorssignificantly greater than success through speculation, t (248) = 7.74, p < 0.001. Thisrepresented a large effect, dz= 0.52. So, the null hypothesis that there is no difference between fundamental factors and speculation in their ability to determine exchange rate isrejected.5.1.1 Importance of Factors Affecting Foreign Exchange Rate over DifferentForecast HorizonsTable 5.3 displays the importance of factors affecting foreign exchange rate overforecast horizons that varies from intraday to more than one year. For the purpose of thisstudy, medium run refers to periods shorter than 6 months and long run refers to periodsmore than 6 months.Table 5.3: Importance of factors affecting foreign exchange rate over different forecast horizonsFor fundamental factors, 52.8 per cent of traders believe that effect of fundamentals isobservable over a period of 1 year, the majority (89.9 per cent) of traders believe that effect is from 6 months to more than a year. Technical factor seems to have an effect over a short period as majority 77.1 percent viewed its importance over intraday and 1 week, 44.8 percent of traders believe that importance of technical factors over a time period of 1week. Like technical, importance of behavioural factors also have importance over a short period, as the majority (86.3 percent) of traders believe that its influence ranges from intraday to 1 week. For speculation, 85.1 percent of traders believe that the effect of speculation is over intraday (very short period).5.2 Analysis of Difference in Success Rate Achieved Through Fundamental and Non Fundamental FactorsIn order to check whether the success rate achieved by foreign exchange tradersthrough fundamental factors differ significantly from the success rate achieved throughnon-fundamental factors, following hypotheses were tested.H04: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and technical factors.H05: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and behavioural factors.H06: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and speculation.To test hypotheses H04- H06, paired sample t-test was performed. The results of pairedsamples t-tests for the difference between the success rate achieved in foreign exchangetrading through fundamental factors and non-fundamental factors have been given intable 4.4 and table 4.5 respectively.Table 4.4 shows the descriptive statistics for hypotheses H04 – H06 i.e. the mean, thenumber of participants (N), the standard deviation of the sample and standard error of mean for each condition and pair. A paired sample t-test is a parametric test based on normal distribution, it becomes necessary to ensure that all the assumptions of the test are satisfied. Although, in large data set (N > 30), the assumption of normality is likely to be satisfied (Due to the central limit theorem). But to prevent large deviation from normality, values of skewness and kurtosis were checked. So, before performing the paired sample ttest, it was assured that assumptions of normality of sampling distribution are fairly met and the data are measured on an interval scale.Table 5.4: Paired samples descriptive statistics for the difference in success rateachieved through fundamental and non-fundamental factors.Skewness values for the difference between the success of factors vary between -1.835and 0.114 which show that the distribution is a little negatively skewed but within the acceptable limits of the normal distribution (+/-2). Thus, the distribution can be accepted as fairly normal. Kurtosis values for the difference between the success of factors vary between -0.668 and 0.732 which indicates that the distributions are little leptokurtic and platykurtic but within the acceptable limits of the normal distribution (+/-2) (Cameron, 2004). So, the distribution can be accepted as fairly normal.Hypothesis H04 compares the success rate achieved by foreign exchange tradersthrough fundamental and technical factors. Success through Fundamental factors (M=6.35, SD= 1.06) found to be greater than through technical factors (M=5.38, SD= 1.26). Hypothesis H05 compares the success rate achieved by foreign exchange traders through fundamental and Behavioural factors. Success through Fundamental factors (M=6.35, SD= 1.06) found to be greater than through Behavioural factors (M=3.54, SD= 1.41). Hypothesis H06 compares the success rate achieved by foreign exchange traders through fundamental and speculation. Success through Fundamental factors (M=6.35, SD= 1.06) found to be greater than through speculation (M=3.61, SD= 1.77). For final confirmation of these results, the difference was checked through pair sample t-test and the same is reported in table 5.5.Table 5.5: Paired samples test for difference in success rate achieved throughfundamental and non-fundamental factorsTable 5.5 shows the results of Paired difference t-test used to analyse the averagedifference between conditions. The results indicate Mean difference, Standard deviationof differences, Standard error of differences, confidence interval at 95% level, the valueof paired t-test, the degree of freedom and 2-tailed level of significance at 5%. Thehypotheses results are as follows:H04: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and technical factors.On average, foreign exchange traders viewed success through fundamental factorssignificantly greater than success through technical factors, t (248) = 8.82, p< 0.05. So,the null hypothesis that there is no difference between the success rate achieved in foreignexchange trading through fundamental factors and technical factors is rejected.H05: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and behavioural factors.On average, foreign exchange traders viewed success through fundamental factors significantly greater than success through behavioural factors, t (248) = 24.65, p < 0.05.So, the null hypothesis that there is no difference between the success rate achieved in foreign exchange trading through fundamental factors and behavioural factors is rejected.H06: There is no difference between the success rate achieved in foreign exchangetrading through fundamental factors and speculation.On average, foreign exchange traders viewed success through fundamental factorssignificantly greater than success through speculation, t (248) = 17.77, p< 0.05. So, thenull hypothesis that there is no difference between the success rate achieved in foreignexchange trading through fundamental factors and speculation is rejected.5.3 Analysis of Difference in the Ability of Behavioural Factors to Influence Foreign Exchange Trading Decisions of TradersBehavioural factors identified from literature includes bandwagon effect, over-reactionto news, market judgements, peer & social influences and rumours. Osterberg and Humes (1993) found that incorrect news and rumours play a substantial role in the foreign exchange market. Allport & Postman (1947) and Rosnow (1991), concluded that importance of the events and amount of doubt involved in the matter increase the occurrences of rumours. In order to check whether the behavioural factors differ in their ability to influence foreign exchange trading decisions of traders, following hypotheses were tested:H010: Behavioural factors do not differ significantly in their ability to influence foreignexchange trading decisions of traders.The results of one way repeated measures ANOVA for the difference in the ability of behavioural factors to influence foreign exchange trading decisions of traders, have been given in table 5.6, table 5.7, table 5.8 and table 5.9 respectively.Table 5.6: Descriptive statistics for behavioural factors ability to influence foreignexchange trading decisions of traders.Table 5.6 shows the descriptive statistics i.e. the mean, the standard deviation of the sample and the number of participants (N) for various constituents of behavioural factors. Over-reaction to news (M= 4.68, SD= 1.56) received highest score followed by Rumors (M= 4.61, SD= 1.72), Market Judgements (M= 4.28, SD= 1.39), Bandwagon Effects (M= 4.04, SD=1.60) and Peer and Social Influences (M= 4.04, SD= 1.48).Table 5.7: Mauchly's test of sphericity for the difference in the ability of behaviouralfactors to influence foreign exchange trading decisions of tradersTable 5.7 indicates that to check the assumption of sphericity, Mauchly’s test was performed. Mauchly’s test is significant (p<0.05) so, the assumption of sphericity has been violated. As the variances of the differences between levels are not equal, two corrections were applied i.e. Greenhouse-Geisser and Huynh-Feldt. Both values of correction, ê (0.87 and 0.89) are closer to 1 rather than lower bound of 0.25. The closer that ê is to 1.00, the more homogeneous are the variances of differences, and hence the closer the data are to being spherical.Table 5.8: Tests of within-subject’s effects for the difference in the ability ofbehavioural factors to influence foreign exchange trading decisions of tradersTable 5.8 shows the results of ANOVA for the within-subject variable. As correction factor was applied to the assumption of sphericity, so, the corrections results (Greenhouse-Geisser and Huynh-Feldt) in the observed F were checked. F-statistic of Greenhouse-Geisser and Huynh-Feldt both are significant, p < 0.001.Table 5.9: Multivariate tests for difference in the ability of behavioural factors to influence foreign exchange trading decisions of traders.Multivariate tests shown by table 5.8 confirms the results of table 5.9 as p < 0.001.These statistics were checked because the assumption of sphericity is satisfied through the correction factor.Mauchly’s test indicated that the assumption of sphericity had been violated, χ2 (9) =66.51, p < 0.05, thus the degree of freedom (df) were corrected using Greenhouse-Geisser and Huynh-Feldt estimates of sphericity (ê= 0.87, 0.89). The results show that the ability of behavioural factors to predict trends in foreign exchange market differs significantly from each other, F (3.51, 863.20) = 15.98, p < 0.001 (Greenhouse-Geisser), F (3.57, 863.20) = 15.98, p < 0.001 (Huynh-Feldt estimates).6. INVESTMENT PORTFOLIO OPTIMIZATION ON RUSSIAN FOREX MARKET IN CONTEXT OF BEHAVIORAL THEORY6.1Development of behavioral aspects in portfolio theoryCurrently, economic theory makes reference to two economic models regarding humanrationality. This is a portfolio theory associated with expected utility [13]. This theory states that the investor is inclined to independently calculate all the risks. The behavioral theory believes that the investor is prone to make mistakes in evaluating information, which leads to losses and incorrect calculation of the probable profit. The behavioral theory was considered in the context of portfolio optimization in finance. Based on this theory, behavioral models were presented for compiling an investment portfolio [17–18].Depending on the type of investor or his investment goal, an approach is used in which he is divided into "rational" and "irrational" parts.Within the framework of this study, it is necessary to determinewhether the use of standard modern portfolio theory is effective. It is also necessary to find out if the behavioralhuman perception errors affect investment portfolio compilation and what techniques can improve portfolio performance.Behavioral models are similar to the interpretation of portfolio theory. Butthe assessment of risk, profitability and utility is different. In behavioral models, objectiveparameters are replaced with subjective ones. Therefore, the resulting utility is perceived not as a product of the risk and return of the asset, but is a subjective assessment of the profitability and the probabilities of its receipt. Quantifying the profitability of an irrational investor differs from the subjective understanding of profitability. There is a distorted understanding of "profit" and "loss".So investors who faced the crisis phenomena of 1998, 2008 and 2014 in Russia want to minimize the risks of investing in savings.Therefore, the Forex market where there is a high level of risk of financial investments and there are no guarantees of investment insurance seems to be unreliable for investment.As a result, investmentsin the Forex Market often began tobe considered solely in terms of the high riskand speculative nature of trading.The lack ofexperience of operations in the Forex Marketand the developing ideas about the risk natureled to a low level offinancial literacy of traders. To overcomethe behavioral effect of avoiding lossesthrough a complete rejection of investment isa pressing issue to improve financial literacy.This led to the fact that investments in the Forex market began to be viewed as unreliable and speculative in nature. In the conditions of financial illiteracy of traders and a lack of experience in the Forex market, distorted ideas about risk have been formed.Today, the problem of improving financial literacy includes, among other aspects, overcoming the stereotype associated with a complete rejection of investments in order to avoid losses.As an example of the study of the irrational nature of human behavior in economics, we can present the article by D. Kahneman and A. Tversky [20], devoted to the presentation of the “theory of prospects”. The author proves the relationship between behavior in case of risk and uncertainty of perspective (probability). People tend to underestimate a situation with uncertain probability and overestimate a situation with precise probability. Then we can talk about the “confidence effect”, which means giving up risk in situations with guaranteed income and looking for risk in situations with guaranteed losses.Moreover, there is“Framing effect” which consists in the fact that the investor does not perceive possible prospects and his choice depends on the wording of the question [21, 22].Preferences are assessed based on the wording of the product or service proposal.Table 6.1: Behavioral portfolio theoriesSource: compiled by the author based on the analysisof references [1, 3, 9, 18, 19, 21].Distorted perception of information in the form of ignoring some alternatives or overestimating one's own confidence leads to an underestimation of reality and an irrational choice. So, there are two key manifestations - errors in simplifying information and overestimating known information. That is, in addition to changing the meaning of utility, estimates of profits or losses, the theory made it dependent on the estimate of probability, and not the level of risk (D. Kahneman and A. Tversky).Later, the theory of prospects by D. Kahneman and A. Tversky [23] was supplemented with the effects of “diminishing sensitivity” and “loss aversion”. The utility function included three main features:1. The function is constructed relative to thegains and losses and a certain “reference point”that refracts it.2. The function has the property of “diminishingsensitivity” which indicates the dependence - the bigger the sums of money are, thesmaller the psychological difference betweenequal intervals of the sums of money.3. The tendency to avoid losses (losses areperceived to be more substantial than gains).The ratio of risk and probability in the extendedtheory included two phenomena:1. Seeking for risk in case of losses and avoidanceof risk in case of gains at high probabilityof loss or gain.2. Seeking for risk in case of gains and avoidanceof risk in case of losses at low probability ofloss or gain.This was due to the fact that people tend tooverestimate low probabilities and underestimatemoderate and high probabilities of events(losses or gains). The functions of the decisionweights are located side by side, but the functionof estimating the probability of gains is alittle more curved than the probability functionof losses. Therefore, avoiding of risk for gains ismore specified than seeking for risk for losses incase of moderate and high probabilities of theseevents (gain or loss).Based on his work, H. Markowitz [13] presentedthe development of the theory for assessingthe profitability of an investment portfolio,which was later called the “Modern PortfolioTheory”.His work was devoted to the analysis of themost optimal investment portfolio in terms ofincreasing profitability and reducing risk. Herefuted the portfolio theory only from the pointof view of maximizing the discounted expectedincome and added dependence on income dispersionto the model, as the investor seeks toreduce the risk of loss of income at the end ofinvestments. According to this theory, a diversifiedportfolio is anyway more preferable for theinvestor than a portfolio without diversification.Compiling a portfolio itself involved two steps: aretrospective assessment of the returns on securities and then an assessment of their potential future returns.For a long time, this interpretation of theasset utility valuation theory in terms of risk /return (CAPM model) was the main practice in compiling investment portfolios. The theory of risk / return assessment through the utility theory was developed in thework by H. Shefrin and M. Statman [3], who consideredthe problem of compiling a portfoliofrom a behavioral point of view. In the developedbehavioral portfolio theory, utility theorieswere analyzed through profitability and riskconsidering the prospect theory and “mental accounting”proposed by R. Thaler [24]. Mental accountingmeant the concept of dividing the numericalexpression of utility in the consumer’smind into separate “accounts”, i. e. Independenttarget sections in the consumer budget. A specificfeature of the “accounts” was their independentassessment in terms of profitability or lossmakingrelative to the past periods. Based onthe theory by psychologistLopes [25] in the SP /the authors propose twoportfolio options: with one and with two mentalaccounts. The portfolio is determined based onthe ratio of probability (risk and level ofexpectationby Lopes) and expected welfare (accordingto H. Markowitz). In case of a portfolio with twomental accounts, it is proposed to add a pyramidalstructure (one account is toaccumulate savingsfor large acquisitions, the other is to savefor a rainy day, with no specific aim). The effectiveboundary line of H. Shefrin and M. Statmanportfolios does not coincide with the effectiveaverage dispersion boundary according to thetheory by H. Markowitz.Subsequently, H. Shefrinand M. Statman [27] applied the criterionproposed earlier [28] when studying the impactof psychological factors on the design and marketingof structured financial products.Later, a generalization of the Markowitztheory with the mental accounting theory wasproposed by S. Das [35]. It was suggested to formulatea portfolio as a whole according to theMarkowitz mean-variance portfolio theory, butto separate the portfolios according to their intendedpurpose (according to the mechanism ofmental accounting) and keep track of each portfolioaccording to the theory by H. Shefrin andM. Statman.The subsequent development of the theory isassociated with the adjusting indicators of thebehavioral model. In the work by S. Das [36], theconcept of risk in the modern portfolio theoryinstead of the standard deviation was replacedby achieving the goal. ResearcherE. De Giorgi [38] showed the importance of theprospect theory instead of the mean-varianceanalysis in the portfolio optimization consideringthe principles of integrated private capitalmanagement (including real and financial assets).Later, E. De Giorgi [39] proposed breaking down the investment process into two stages:setting goals and investing for each goal accordingto a specific strategy by separating the goalsinto short-term and long-term ones. Moreover,E. De Giorgi and S. Legg [17] applied the modelby N. Barberis and M. Huang [40] using “narrowframing” and “loss avoidance” to constructa mathematical portfolio model. E. De Giorgi[41] also gave a mathematical formulation ofthe “naive diversification” model (that is, thephenomenon of preference for uniformdiversificationbetween all assets or preferences of acertain type of known assets). Changes in the parameters of the behavioralportfolio theory are presented in table 1.As it is seen from table 1, the introduction ofbehavioral factors in the portfolio theory tookplace in stages. Certain aspects of the behavioraltheory were used to justify such effects as lossaversion, framing (dependence of the perceptionof information on its presentation form), mentalaccounting, and behavioral finance phenomenasuch as a section of behavioral economics thatstudies behavioral effects in the stock market(for example, “ naive diversification”). In thiscase, the first stage of implementation was characterizedby a key replacement of indicators withbehavioral, estimated values, and subsequentlythe behavioral theory was considered as anintegralpart of the general portfolio optimizationtheory.The study of the behavioral effects in theRussian market was based on the identificationof the phenomena already seen in the US stockmarket. Therefore, the behavioral finance wasdeveloped to a greater extent. Thus, V. R. Evstigneevconsidered the decision-making mechanismin the foreign exchange market basedon the expectations of other participants [42].Decision models in the foreign exchange marketbased on the Bayesian procedure were alsoproposed by Yu. V. Yeltsov [43]. In the securitiesmarket V. R. Evstigneev proposed toformalizethe cognitive dissonance effect through thematrix operator of the observed securities yieldvector [44].Also V. R. Evstigneev proposed a predictionmodel by an investor based on a predictablerandom process leading to the rejection ofmaximizing utility in favor of attempts to hitthe jackpot in each individual case [45]. In herwork, V. A. Goretskaya noted the importanceof applying the prospect theory as the basis ofthe behavioral finance for decision-making inthe stock market [46]. The issues of informationasymmetry in the financial market of Russiawere also considered by V. P. Ivanitsky andV. A. Tatyannikov [47].Therefore, it is necessary to assess the impact of behavioral errorsprivate investor who tends to overestimatethe likelihood of losses and profits when drawing upinvestment portfolio.6.2 Comparison of average risk, return and portfolio quality in behavioral modelsThe field of study of finance at the present stageincludes a set of theories that are not supported by practical tika, and practices that are not explained by existing theories. Noting that trade in financial markets still billions of dollars today ditch, experts argue that management companies in areas of wealth management are increasingly using to using the concept of behavioral finance in their work, seeking to restore the confidence of their clients. This can be confirmed by the fact that several leading world financial institutions at once mulberries, including Merrill Lynch, Northern Trust and JP Morgan. Chase are actively implementing behavioral strategies today money into their day-to-day business, although everyone of them does it differently.Members of global finance owl communities at all levels agree, which is one of the most profound consequences of the currentcrisis has become the increasing importance of various "Emotional factors" in the decision-making processni by investors owning invested assets in the amount of $ 1 million.Developed in the 1970 and 1980 scientists-the-retics, including Amos Tversky, DanielKahneman, Richard Thayer and Meir Statman, the theory of financial finance is based on the postulate that in certain moments of psychology and emotions make investors behave irrationally from the point of view modern portfolio investment theory.Concept behavioral financial showing that the market irrational behavior of agents is often used, which is impossible to explain the point of view classic finance. At the same time, there are four types of effectiveness of arbitration associated with the irrationality of market participants:fundamental risk, institutional constraints, the risk of irrational traders, exposure of professional traders to universal human shortcomings [12, 14].Thisthe concept is today developed to the level ofthat the increased use of behavioralfinance in the investment process will provide significant impact on delivery models and platforms client services. Management companies in the area wealth management can get competitive advantage in the market due to the inclusion of the concepttions of behavioral finance into their investment strategy as it improves the quality of management portfolio and risk assessment models, and expands their capabilities.The growing interest in behavioral finance can bebe regarded as a paradigm shift in the consideration of the problem formation of the structure of investment portfolios. Behavioral finance exposes flaws and failures compliance with the theory of modern investment portfolio when used in relation to realcustomers in the real world, opening up the possibility of creatingDenmark a more efficient asset allocation model in the investment portfolio. As a consequence, the global the financial market today is at a strategic the stage of development from which the movement from modern theory of investment portfolio in our reign of a new era of targeted planning.The concept of loss aversion, inwithin which investors perceive the risk of receivinglosses on their investments as unacceptable, considered is the determining factor in this transition to a new paradigm. On the basis of this concept, most number of behavioral models for portfolio optimization the securities market.In contrast to the theory of modern investment portfolio proceeding from symmetric dependence between profitability and risk, taking into account psychology markets overlapping behavioral finance reveals the presence of asymmetry between the size of the investment positional profit and the level of risk. When "rejection losses "is taken into account, the models are used more than consciously.Mostly Irrational Market Behavioris the result of insufficient or excessive the reaction of investors in relation to a particular financial instrument. This the idea works this way: if an analyst or an individual dual investor come to the conclusion that insufficient or the overreaction is irrational, they buy or sell this asset against market in order to take advantage of the “cost an anomaly. " An example of overreaction would be serve a strong fall the market capitalization of BP-related companies, after the accident on the deepwater platform Deepwater Horizon and oil spill in the Gulf of Mexico.Noise based tradingopposed to trade based on timely accurate and accurate information. If market participants use unverified data and rumors, advice "Experts", in fact, are not, They carry out trade transactions, mistaking "noise" for information, not having real information or simply because, in the words of F. Black, "they like to trade." By-their maintenance is irrational. Precisely the presence in the market of the element of "noise trading" determines the possibility of a market and implementation trade transactions. If market prices were the result of adequate information, the market cannot you could get additional profit, trade would lose its meaning [4, 8].In accordance with this theory, the market is also obliged torational participants must also be present, the so-called "information traders". If a consider the market as a whole, "noise traders" turn out to be at a loss, while “information traders ”make a profit. Moreover, the latter intheir actions take into account, of course, and behavior "Noise", irrational market participants and trade to a greater extent with them than with the same as they, "Information traders".The assumptions of the theory of noise trading contradictchat to almost all the most important theories of classicalfinance. But stock practice fits within its framework and everything that happens in the market, although the behavior itself the market turns out to be practically unpredictable.In the traditional problem of asset allocation inportfolio, the level of risk acceptance, constraints and financial financial goals are set once, the result tats are given by mean-variance optimization.Unfortunately this approach falls short credibility for investors, receptive to behavioral deviations (bias). To at-measure due to speculative market movements and damage to long-term investment plans individual investors may revise the port fel. This is an indication that an early transfer of assets in cash can cause momentum (technical indicator).With regard to emotional biases, empirical studies have shown that when it comes to about profit, the investor is not inclined to take risks, but when alternatives include losses, actions of investor are too risky (asymmetric behavior risk-related investors). With a wider people are much less disposed to choose assets to losses, they are less attracted by the profit of the same sizes. Loss aversion is an essential psychological concept that has been adapted tofinancial and economic analysis and now provides the basis for the theory of perspectives.Dominant the paradigm of individual behavior in financial oriya is based on the postulates of maximizing the expected the usefulness and risk aversion that are recently criticized for obvious mistakes.Practical psychology demonstrates that people withsystematically deviate from the predicted choice, which the classical theory predicts, since individuals are typically biased in their perception real facts. Based on market research USA and Canada, foreign scientists have established statistics significant correlation between the negative psychological qualities of a traderand the success of his professional activity. In accordance with this theory, the key to success at the market is the presence of a potential investor following features:- lack of desire to subjugate the market and controll him;- the ability to feel the individual "barrier risk ", i.e. limiting values ​​of the magnitude of risk, which investor can take upon himself, and the maximum the value of capital that he can risk without fatal consequences;- the ability to act outside the zone of psychological comfort, i.e. take deliberate and adequate solutions, even if the situation does not develop as the investor assumed at the beginning;- the ability to recognize the state of stress and expression fight defense mechanisms against rashactions in this state, to distance yourself from personal emotions and experiences;- the investor has adequate self-esteem;- the ability to take into account the influence of negative attitudes;- the ability to overcome psychological attachment; -sensitivity to specific financial instruments;- the ability to abandon momentary profits for the long term;- the presence of endurance and patience;- the ability to plan several (often the opposite positive) options for the development of events in the stock market;- the ability to focus on the goal and constantly act on the basis of the decision;- have the skills to work with large arrays of information formation in order to eliminate psychological changes load;- lack of psychological dependence on tradinga.The listed qualities allow you to copewith negative consequences of errors of perceptioninformation and emotional factors that cause deviations from rational behavior in the market.A portfolio formed by an investor whose choice consistent with prospect theory, will be significantly differ from portfolios of traditional (rational) investors whose choice is made in accordance with the theory expected utility. In general, the nature of behavioral portfolios are determined by their ability to combine very reliable securities with very risky nym. In this vein, the optimal solution to the problem portfolio choice should be the one that is most to a greater extent meets their interests.It shouldbe prescriptions or rules for portfolio selection, which correspond to natural preferences and the investor's mindset (his emotional inclination stam), even if this approach does not maximize expected receipts at a given level of risk.This does not mean that the individual investor is irrational:it is not irrational for people that they have certain emotional reactions, take them taken into account in the decision-making process and try bring your choice in line with your preferences readings or attitudes.Nevertheless, portfolio managers are testinglack of modern techniques required to combine these tendencies in the process of placement niya finance in assets.In general terms, perspective theory and its subsequentversion - cumulative perspective theory – postulate four new concepts in the system of individual risk preference. The first is that investors assess the share tives according to gains and losses, and not according to the final wealth (mental accounting). The second is that individuals are they are disposed to losses, which is why they attract gains (loss avoidance). The third is that individuals are more are risk-averse and risk-averse in the area of ​​winnings (asymmetry of risk preferences).Fourth, individuals estimate marginal probabilitiesin a form that overestimates low probabilitiesand underestimates high values ​​(function of weightedprobabilities).Based on the above, we can conclude thatthat a behavioral risk exists. One side,its nature is determined by human psychology and individual perception of risk. On the other hand, this feature affects the level of investment risk and results.6.3Investment portfolio optimization modelModern researchers are trying to overcomethe problem of classical models of portfolio management that cannot be created in real conditions.After reviewing the relevant material, the author expects resampling techniques to offer typically better results.It introduces a new approach to implementing behavioral propensities and approximate risk assessment into variance portfolio choice (adjusted technique behavioral resampling). The result was a model is proposed for assessing portfolio choices with loss avoidance, with asymmetric risk-taking behavior and separation of risk-free opportunities.The data obtained suggested thatchanges in portfolio weights are critically dependent onfixed reference point, the relationship between this point and current wealth and thus indirectly depend on the behavior of risky assets.Such a model evaluates non-the effective cost of behavioral tendencies and offer a form for calculating the profitability of risky assets, close to reality, including an approximate risk assessment in analysis. When analyzing the results of day trading on highly liquid markets, such a problem, as a rule, does not arise - a large the number of transactions guarantees the availability of price information, while the result of each individual transaction does not significantly affect the average.The situation changes dramatically when analyzing low-liquid markets or when studying the microstructure of markets carried out onbased on intraday information. In both cases, periods inevitably occur during which there are no market transactions.At the same time, in short-term fluctuations in market liquidity or in the case of non-market transactions, their price may be differ from the prices of transactions that took place earlier, as well as go beyond the framework of the established quotes for buying and selling.The presence of both phenomena - low-quality information and omissions data - leads to inadequate results when using numerical procedures, and sometimes to the impossibility of their application. The solution to this problem can be carried out in the following way: by taking into account in the model the possibility of the presence of a missing data and transactions that stand out against the background of the main amount by atypical values ​​or volumes. In this case, we are talking about the application of the method of repetition sample by restoring data.The data recovery process will be demonstrated by the most complete set of information describing the results of bond trading on the Moscow Interbank Currency Exchange. Information recovery results and methods of their application can be different. Often, as in solving the problem of constructing a bond yield curve, it is enough to obtain a point estimate of the missing values ​​and the accuracy characteristics (figure 6.2)..Figure 6.2: Initial price values ​​and recovery resultsmissed information on trading in bonds "Russia 25057"for the period 30.10.2017-17.01.2018The solid line shows the actual values ​​of prices, the thick dashed line is the estimate of the average price value using the Monte Carlo chain, the thin dashed lines show the boundaries of the 10% confidence interval.Fig. Figure 6.2 illustrates the typical behavior of the individual bond price process and the results of data recovery. The average price value is obtained as the arithmetic average of the results of the iterations. The upper and lower limits of the confidence interval are sample quantiles with significance levels of 0.05 and 0.95, respectively.An example of the distribution of results for one separately taken missing value is shown in Fig 4. It is possible to determine the quality of the estimate in terms of the ability of the algorithm to reproduce the unknown values ​​of the parameters only when working with a distribution with previously known characteristics. For this purpose, a number of numerical experiments were carried out.As an initial process, we used a multidimensional geometric Brownian motion with parameters that more or less plausibly reproduce the dynamics of market processes. The elements of this process were eliminated at random, followed by the restoration of the joint distribution of parameters and missing data. Fig. 6.3, 6.4 illustrates the quality of the estimates obtained from the point of view of the accuracy of retrieving the process parameters.Figure 6.3: Empirical distribution of the restored pricebonds "Russia 25057" for November 7, 2017Figure 6.4: Joint empirical distribution of mean values ​​of two componentsThe bold round dot shows the real value of the parameters. Bold square point - estimation of the same parameters using the EM algorithmFigure 6.5: Histogram of the empirical distribution of the mean valueThe solid vertical line shows the real value of the parameter. The dashed vertical line shows the position of the parameter estimate using the EM-agglomeration.Fig. 6.5 gives an idea of ​​the accuracy of missing data recovery. It should be noted that the estimates obtained in most cases lie in the zone of the highest concentration of the obtained empirical distributions. According to the results, the real values ​​of the parameters never went beyond the 95% confidence interval of the obtained empirical distributions.At the same time, the parameters of price distribution were estimated and the missing data was restored using the EM-algorithm. The comparison results of the methods used can be interpretedtwofold.On the one hand, the EM-algorithm gives estimates of the parameters of the distribution of the process, which coincide with the modes of the empirical distribution of parameters obtained when constructing a Markov Chain Monte Carlo Methods. On the other hand, the estimates of the values ​​of the obtained values ​​differ significantly both from their real values ​​and from the estimates obtained by reconstructing the distribution (see Fig. 6.6). At the same time, there is a very slow dynamics of changes in the estimates of the values ​​of the restored missing data, as a result of which the reconstructed time series is smoothed, and the estimates of price volatility obtained on the basis of such data are underestimated. The estimates obtained using the Monte Carlo chain with the Markov property are free of the indicated drawbacks.Figure 6.6:Empirical distribution of the values ​​of the covariance coefficientsOne more advantage of the data recovery method using the Markov chain should be noted. The results obtained can be directly used to obtain the distribution of the value of the portfolio of assets in the future.7.FORECASTS AND APPROPRIATENESS7.1ResultsSo Ultimately, a new approach to portfolio selection was developed in order to visualize an approximate error estimates in the traditional optimization method portfolio more convenient for investment analysts. Unified method of selection from multidimensional normal distribution (parametric method, called Monte Carlo simulation) in our standing time is the way to solve the problem estimates of the error in choosing the optimal portfolio. Comparison of the results of applying the traditional approach to resampling and this approach to make conclusions about the success of the latter.Researchers are evaluating several patterns ofassignments of assets and assume that resampling methods offer typically better results. They present a novel approach to integrating behavioral tendencies and risk estimates into variance portfolio selection (adjusted techniquebehavioral resampling).7.2 Interpretation, Discussion, Future prospectsBased on reviewed studies and allnew aspects of theories and strategies can be created behavioral models for optimizing portfolio selection. Such models estimate the ineffective value of behavioral propensities and offer a form for calculating the profitability of risky assets, close to reality, including an approximate risk assessment in analysis.Using this and other empirical data,the founders of the behavioral school tried to describe how decisions are made in reality, determining the capital structure. The result wasa number of behavioral theories have been created that have revealed interesting patterns of behavior in questions capital structure.These theories show that optimism and overconfidence is significant factors that determine the capital structure, which no traditional theory. It doesn't matter how the behavioralfinance finds its application in investment strategies in practice. It is quite clear that this the concept has finally and irrevocably entered our a life. The use of behavioral finance has become a global socio-economic trend that will create significant opportunities for growth and transformation of this industry.CONCLUSIONThe foreign exchange market is the largest highly and liquid financial market in the world with worldwide average daily turnover around $5.3 trillion, which makes foreign exchange highly global trading asset (Rime & Schrimpf, 2013). Foreign exchange forms the basis of dealings for trade and other monetary transactions between economies of the world. Trading participants in this market include central banks, commercial banks, companies, brokers, fund managers, speculators and individuals. Every participant aims to achieve its objectives with the best possible management of risks to which one is exposed. High liquidity comes with high volatility which makes management of risk difficult. Since the global capital is highly volatile and forms its basis from country fundamentals and investor sentiments. This volatility influences exchange rate and in turn the trade balance and cost of goods.The main rolein the formation of the exchange rate "Moscow Exchange", one of the key shareholders of which is Bank of Russia. Today the MB is the largest exchange in Russiaholding, one of the twenty leading world sites in terms of volume trading in securities and foreign exchange resources. As is well known, in financial and economic sphere, the IB is the undisputed leader among other Russian stock exchanges, accounting for about 70% of the exchange turnover for the US dollar.Total trading volume on the Moscow Exchange markets in 2020amounted to 947.4 trillion. rubles.All major exchange markets demonstrated positive annual dynamics of the trading volume: the stock market (an increase of 97.3%), the derivatives market (an increase of 57.7%), the money market (an increase of 23.2%), the bond market (an increase of 8 , 5%) and the foreign exchange market (an increase of 6.7%), including the foreign exchange spot market (an increase of 43.9%)The nominal exchange rate is subject to scrutiny.attention of business entities and the state, since its stability characterizes the stability of the national currency, necessary for progressive economic development and ensuring welfare of the country's population.Currently in the system of financial and economic relationscurrency relations are relations associated with the use of currency in foreign trade, in the implementation of foreignborrowing, attracting foreign investment, making transactions when buying or selling currency, with a number of banking operations with currency, customs movement of currency and currency values. So Thus, this is a very relevant topic in the context of development and globalization national economy, which has not only advantages, but also shortcomings with which states are actively fighting.In the next few years, the Russian foreign exchange market will be more or lessstable, however, the renewal of the BR policy regarding the purchase of foreign currency may lead to the development of some "deviations" from the predicted course.Also, according to experts, Russia has already been able to overcome the conditionseconomic crisis and stabilized its economy so much that in next year there will be no currency jumps.Impact of several factors such as economic forces, market psychology, political influences and availability of several technical trading rules etc. makes the prediction of the exchange rate a highly volatile and intricate phenomenon. Even forex traders, academicians and economists who study the forex market every day find it difficult to accurately predict exchange rate movements (Hill, 2012). All market participants whether central banks, commercial banks, multinational corporations or brokers, speculators and individuals etc. make attempts to accurately predicting future exchange rate movements in order to satisfy their respective objectives.Behavioural finance research has contributed significantly in the field of foreignexchange trading by exploring through different methods factors that affect trader’s psychology. Behavioural factors comprise variables that affect a trader’s psychology such as bandwagon effects, overreaction to the news, market judgment, peers & social influences and rumours etc. Overreaction to news followed by bandwagon, effects are the most important driving factors that cause movements in the exchange rate (Cheung, Chinn, & Marsh, 2004). Successful forex trading requires a total approach that integrates fundamentals, technicalanalysis, and psychology (Cofnas, 2012).The field of study of finance at the present stageincludes a set of theories that are not supported by practice and practice that is not explained by existing theories. Net present value rule, fundamental investment theory, efficient stock market theory, modern portfolio theoryinvestment, CAPM model, synergetic theory mergers and acquisitions - these are just a few theories, whose practical significance every day inspires everything more and more serious concerns.The field of study of finance at the present stageincludes a set of theories that are not supported by practical tika, and practices that are not explained by existing theories. Noting that trade in financial markets still billions of dollars today ditch, experts argue that management companies in areas of wealth management are increasingly using to using the concept of behavioral finance in their work, seeking to restore the confidence of their clients. This can be confirmed by the fact that several leading world financial institutions at once mulberries, including Merrill Lynch, Northern Trust and JP Morgan. Chase are actively implementing behavioral strategies today money into their day-to-day business, although everyone of them does it differently.The modern economic theory provides twopoints of view on the premise of human rationalityin economic models. According to theexpected utility theory, on which the modernportfolio theory is based, the investor is inclinedto independently calculate all the risks.The behavioral theory followers in economicsbelieve that an investor tends to make mistakesin evaluating information, probabilities, and estimatinglosses and gains. The behavioral theory,in particular, was considered in the context ofportfolio optimization in finance. On this basis,behavioral models were presented to compile aninvestment portfolio.The current investment portfolio models usean approach where it is divided into “rational”and “irrational” parts, depending on the typeof the investor or his investment goal.However,the questions remainThe growing interest in behavioral finance can bebe regarded as a paradigm shift in the consideration of the problem formation of the structure of investment portfolios. Behavioral finance exposes the flaws and inconsistencies of the theory of modern investment portfolio when used in relation to real clients in the real world, opening up the possibility of creating a more efficient asset allocation model in the investment portfolio. As a consequence, the globalthe financial market today is at a strategic the stage of development from which the movement from modern theory of investment portfolio towards a new era of target planning.The author believes that the approach to portfolio selection is effective in order to visualize approximateerror estimates in the traditional optimization method portfolio. Unified method of selection from multidimensional normal distribution (parametric method, called Monte Carlo simulation) in our standing time is the way to solve the problem estimates of the error in choosing the optimal portfolio. Comparison of the results of applying the traditional approach to resampling and this approach to make conclusions about the success of the latter.The results of the proposed methods can be used both as input information for other computational procedures (for example, for constructing a risk-free yield curve), and directly, for example, for solving the problem of assessing the Value-at-Risk indicator of a portfolio. The results obtained can serve as confirmation of the practical significance of this approach.BIBLIOGRAPHYBernstein J. The investor’s quotient: the psychology ofsuccessful investing in commodities & stocks. New York: Wiley, 1993.Beine, M. and S. Laurent (2003) "Central Bank Interventions and Jumps in Double LongMemory Models of Daily Exchange Rates," Journal of Empirical Finance 10(5),641-60.Black F. Noise // The journal of fi nance. 1986. Vol. 41.Bhanumurthy, N. R. (2006). Macroeconomic fundamentals and exchange rate dynamicsin India: Some survey results. Economic and Political Weekly, 1101-1107.Dominguez, K. (1998) "Central Bank Intervention and Exchange Rate Volatility,"Journal of International Money and Finance 17(1), 161-190.Dominguez, K. M. (2003). The market microstructure of central bankintervention. Journal of International Economics, 59(1), 25-45.De Long, Bradford J., Shleifer A., Summers L. Noise traderrisk in fi nancial markets // Advances in behavioral fi nance. New York:Russell Sage Foundation, 1993.Jorion P. Bayes-Stain estimation of portfolio analysis // Thejournal of fi nancial and quantitative analysis. 1986. № 21 (3).Horst J., Roon F., Werker B. Incorporating estimation riskin portfolio choice // Center working paper. URL: http://ssrn.com/abstract=244695.Kempf A., Kreuzberg K., Memmel C. How to incorporateestimation risk into Markowitz optimisation. Operation researchproceedings, 2001.Maenhout P. Robust portfolio rules and asset pricing // Reviewof Financial Studies. 2004. № 17 (4).Markowitz H. Portfolio selection. The Journal of Finance.1952;7(1):77–91.Martin, A. D. (2001). Technical trading rules in the spot foreign exchange markets ofdeveloping countries. Journal of Multinational Financial Management, 11(1), 59-68.Odean T. Are investors reluctant to realise their losses // Thejournal of fi nance. 1998. №53 (5). Oberlechner, T., & Hocking, S. (2004). Information sources , news , and rumors infinancial markets : Insights into the foreign exchange market, 25, 407–424.De Giorgi E. G., Legg S. Dynamic portfolio choice and asset pricing with narrowframing and probability weighting. Journal of Economic Dynamics and Control. 2012; 36(7):951–972.