CC BY
139
ВОПРОСЫ УПРАВЛЕНИЯ
COMPARING PROFITABILITY AND RANKING OF TECHNICAL ANALYSIS INDICATORS BASED ON TOPSIS TECHNIQUE
Ghobadi M.
M.Sc. in Financial Engineering, Department of Financial Engineering, Dehaghan Branch, Islamic Azad University (Iran),
Ghobadi M.
Ph.D. Student of Economic Science, Department of Economic, Isfahan (Khorasgan) Branch, Islamic Azad University (Iran),
Heshmatpour M.
M.A in Economic Science, Tarbiat Modares University, Tehran (Iran), [email protected]
УДК 33:519
ББК65в631
The present research is about investment effectiveness ranking based on technical analysis indicators by TOPSIS technique. This research is kind of experimental post facto research. sample data Include copper, palladium, oil, gold, silver, wheat, sugar and dollar index "between beginning of 2008 to the end of 2013". When the investment strategies are evaluated from above aspects, it can be regarded as a multi-criteria decision-making (MCDM) problem. In addition, it presents a non-additive Technique for Order of Preference by Similarity to Ideal Solution to evaluate technical indicators. Consequently, By ranking TOPSIS Technique performance values, we can determine the relative importance of criteria and make the best investment decision.
According to results all the technical analysis indicators which used in research can find profitable trading prices and all the returns are more than zero. The summary of results for this sub-period describe that in row, STO with a mean of (0.059), RSI with a mean of (0.055), CCI with mean of (0.052), MACD with a mean of (0.048), and MFI with a mean of (0.046) everyone has an abnormal return greater than the risk-free interest rate with a mean of 0.004.
Key words: Technical Analysis Indicators, TOPSIS Technique, Profitability Ranking.
Introduction
The complexities are numerous, and overcoming these g complexities to offer successful selections is a technical analyst's challenge. It is important that the limited amount § of investing Portfolio should be efficiently allocated over o many stocks. The technical analysts need to forecast future "is prices to reduce the risks and find optimal combination of -c optimal indicators out of many technical indicators. The 3; purpose of technical analysts is maximizing the returns g- in allocating indicators Importance to many indicators. In ^ a theories problem, the solution of the portfolio selection -§ problem presented by H. Markowitz (1952) has a tendency ;§ to increase the number of stocks selected for investors.
This paper explores which indicator, including the Rela-^ tive Strength Index; Stochastic Oscillator; Simple Moving -§ Average; Money Flow Index; Commodity Channel Index ;§ has optimal trading ability can lead to high financial perfor-© mance. The financial performance is evaluated by TOPSIS
multi-criteria decision-making (MCDM); this information could support technical analysts' decision - making.
In real investment systems, the decision-making problems are very often uncertain or vague in a number of ways. This type of uncertainty has long been handled appropriately by probability theory and statistics. However, in many areas of financial problems, such as investment management, market microstructure, financing and others decisions often employ natural language to express thinking and subjective perception.
Multi-criteria decision-making forms an important part of the decision process for both the small (an individual) and the large (an organization) investment. When available financial information is precise, many methods exist to evaluate the investment. The methods used to analyze securities and make investment decisions fall into two very broad categories: fundamental analysis and technical analysis. Fundamental analysis involves analyzing the characteristics of a company in order to estimate its value. Technical
Ghobadi M., Ghobadi M., Heshmatpour M.
analysis takes a completely different approach; it does not care one bit about the "value" of a company or a commodity. Technicians are only interested in the price movements in the market. Technical analysis is a method of evaluating securities by analyzing the statistics generated by market activity, such as past prices and volume. Technical analysts do not care whether a stock is undervalued - the only thing that matters is a security's past trading data and what information this data can provide about where the security might move in the future. These days by improvement in financial and computer Technology one of the ways that investors extensively use is technical analysis. One of advantage of Technical analysis is finding the best trading price. Finding these trading prices, help the investors that trade timely and consecutively in the markets by increasing and decreasing the prices speculation and in a period earn abnormal returns.
Theoretical and background research
Technical Analysis: The Technical Analysis origins back to Charles Dow research in early twentieth on industrials Dow Jones Averages index. His research on the market caused him realize that market trends and speculations do not simply follow financial statements and information and there are some other factors that affect the market. Those researches made a great development on the markets price forecast methods. Dow Theory made by collection of his articles in Wall-street journals "between (1851) to (1902)". The Dow Theory on stock price movement is a form of technical analysis that includes some aspects of sector rotation and his theory mainly focused on market trends Dow Theory basis conclude that the fact prices are affected by all the information and events in the markets. All the available knowledge to market participants, including investors or fund managers affects in the price. Early attempts in academic councils assess the effectiveness of technical analysis considered very simple rules called filter rules. These rules involve buying a security if it had increased by x% on the last period or selling it if its price has decreased by x% on the last period.
Topsis
The Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) is a multi-criteria decision analysis method, which was originally develop by Hwang and Yoon in 1981 with further developments by Yoon in 1987 and Hwang, Lai and Liu in 1993. TOPSIS is based on the concept that the chosen alternative should have the shortest geometric distance from the positive ideal solution and the longest geometric distance from the negative ideal solution. It is a method of compensatory aggregation that compares a set
of alternatives by identifying weights for each criterion. Normalizing scores for each criterion and calculating the geometric distance between each alternative and the ideal alternative, which is the best score in each criterion.
An assumption of TOPSIS is that the criteria are monotonically increasing or decreasing. Normalization is usually required as the parameters or criteria are often of incongruous dimensions in multi-criteria problems.
Ghobadi (2014) test the Profitability of Technical Analysis Indicators to Earn Abnormal Returns in International Exchange Markets from 2008 through 2013. They conclude that the positive returns according to technical analysis indicator returns and these returns is significantly more than London Interbank Offered Rate. They observe that the Stochastic Oscillator, Relative Strength Index, Money Flow Index, Commodity Channel Index, Simple Moving Average indicators produces the best results, followed by the London Interbank Offered Rate.
Ghobadi (2014) in a his thesis about "Profitability of Technical Analysis Strategy to Earn Abnormal Returns in TSE (2007-2013)" report abnormal returns using technical trading strategy in the Tehran Stock Exchange by statistical tests. They note that Technical Analysis can signal optimal trading prices and give Abnormal returns more attractive than Central Bank of The Islamic Republic of Iran Risk Free Interest Rate.
Lehmann and Modest (1987) combined the APT performance evaluation method with the Treynor and Mazuy (1966) quadratic regression technique. They found statically significant measured abnormal timing and selectivity performance by mutual funds. They also examined the impact of alternative benchmarks on the performance of mutual funds finding that performance measures are quite sensitive to the benchmark chosen and finding that a large number of negative selectivity measures. In addition, Henriksson (1984) found a negative correlation between the measures of stock selection ability and market timing. Lee and Rahman (1990) empirically examine market timing and selectivity performance of mutual funds. It is important that fund managers be evaluated by both selection ability and market timing skill.
In 1974, Sugeno introduced the concept of fuzzy measure and fuzzy integral, generalizing the usual definition of a measure by replacing the usual additive property with a weak requirement, i.e. the monotonic property with respect to set inclusion. In this section, we give a brief to some notions from the theory of fuzzy measure and fuzzy integral.
Ben R Marshall and Jared M. Cahan (2006) evaluated the profitability of CRISMA technical trading system. They collect information of companies on CRSP database in the period of January1, 1976 to December31, 2003 including 200 days of past closing prices and 20 days of past volume. They examine both long and short CRISMA filter rules in this study and found that even the system generates some profit but not consistently.
Ghobadi M., Ghobadi M., Heshmatpour M.
Research Hypothesis
1. Technical analysis indicators can forecast profitable investment.
H0: ^ > 0
H1: ^ < 0
2. From Topsis technique point of view, which technical analysis indicator returns priority is higher?
H0: = = ... = i.e., all indicator return means are equal.
HA: At least two of the means differ.
Variables
Relative Strength Index
1. RSI = 100-
100 1 + RS
2. RS = Average Gain / Average Loss
3. Average Gain = [(previous Average Gain) X 13 + current Gain] / 14
4. Average Loss = [(previous Average Loss) X 13 + current Loss] / 14
The Relative Strength Index (RSI) is a momentum oscillator that measures the speed and change of price movements. RSI oscillates between 0 and 100. Traditionally, and according to Wilder, RSI considered overbought when above 70 and oversold when below 30. Signals can also be generate by looking for divergences, failure swings and centerline crossovers. RSI can also use to identify the general trend. RSI considered overbought when above 70 and oversold when below 30. These traditional levels can also adjust to fit the security or analytical requirements. Rising overbought to 80 or lowering oversold to 20 will reduce the number of overbought/oversold readings. Short-term traders sometimes use 2-period RSI to look for overbought readings above 80 and oversold readings below 20.
Stochastic Oscillator (STO)
1. %K = (Current Close - Lowest Low) / (Highest High - Lowest Low) X 100
2. %D = 3 - day SMA of %K
3. Lowest Low = lowest low for the look - back period
4. Highest High = highest high for the look - back period
5. %K is multiplied by 100 to move the decimal point two places
Developed by George C. Lane in the late 1950s, the Stochastic Oscillator is a momentum indicator that shows the location of the close relative to the high-low range over a set number of periods. According to an interview with Lane, the Stochastic Oscillator "doesn't follow price, it doesn't follow volume or anything like that. It follows the speed or the momentum of price. As a rule, the momentum changes direction before price." As such, bullish and bearish divergences in the Stochastic Oscillator can be used to foreshadow reversals. Lane identified this first, and most important, signal. Lane also used this oscillator to identify bull and bear set-ups to anticipate a future reversal. Because the Stochastic Oscillator is range bound, is also useful for identifying overbought and oversold levels. The Stochastic Oscillator measures the level of the close relative to the high-low range over a given period. Assume that the highest high equals 110, the lowest low equals 100 and the close equals 108. The high-low range is 10, which is the denominator in the %K formula. The close less the lowest low equals 8, which is the numerator. 8 divided by 10 equals .80 or 80%. Multiply this number by 100 to find %K %K would equal 30 if the close were at 103 (.30 x 100). The Stochastic Oscillator is above 50 when the close is in the upper half of the range and below 50 when the close is in the lower half. Low readings (below 20) indicate that price is near its low for the given time period. High readings (above 80) indicate that price is near its high for the given time period. The IBM example above shows three 14-day ranges (yellow areas) with the closing price at the end of the period (red dotted) line. The Stochastic Oscillator equals 91 when the close was at the top of the range. The Stochastic Oscillator equals 15 when the close was near the bottom of the range. The close equals 57 when the close was in the middle of the range.
Simple Moving Average (SMA)
1. Daily Closing Prices: 11, 12, 13, 14, 15, 16, 17
2. First day of 5-day SMA: (11 + 12 + 13 + 14 + 15) / 5 = 13
3. Second day of 5-day SMA: (12 + 13 + 14 + 15 + 16) / 5 = 14
4. Third day of 5-day SMA: (13 + 14 + 15 + 16 + 17) / 5 = 15
A simple moving average is formed by computing the average price of a security over a specific number of periods. Most moving averages are based on closing prices. A 5-day simple moving average is the five-day sum of closing prices divided by five. As its name implies, a moving average is an average that moves. Old data is dropped as new data comes available. This causes the
Ghobadi M., Ghobadi M., Heshmatpour M.
average to move along the time scale. Below is an example of a 5-day moving average evolving over three days. The first day of the moving average simply covers the last five days. The second day of the moving average drops the first data point (11) and adds the new data point (16). The third day of the moving average continues by dropping the first data point (12) and adding the new data point (17). In the example above, prices gradually increase from 11 to 17 over a total of seven days. Notice that the moving average also rises from 13 to 15 over a three-day calculation period. Also, notice that each moving average value is just below the last price.
Money Flow Index
1. Typical Price = (High + Low + Close) / 3
2. Raw Money Flow = Typical Price X Volume
3. Positive Money Flow = Sum of positive Raw Money Flow over 14 periods.
4. Negative Money Flow = Sum of negative Raw Money Flow over 14 periods.
5. Money Flow Ratio = (Positive Money Flow) / (Negative Money Flow)
6. Money Flow Index = 100 - 100 / (1 + Money Flow Ratio)
The Money Flow Index (MFI) is an oscillator that uses both price and volume to measure buying and selling pressure. Created by Gene Quong and Avrum Sou-dack, MFI is also known as volume-weighted RSI. MFI starts with the typical price for each period. Money flow is positive when the typical price rises (buying pressure)
and negative when the typical price declines (selling pressure). Typically, MFI above 80 is considered overbought and MFI below 20 is considered oversold. Strong trends can present a problem for these classic overbought and oversold levels. MFI can become overbought (>80) and prices can simply continue higher when the uptrend is strong. Conversely, MFI can become oversold (<20) and prices can simply continue lower when the downtrend is strong. Quong and Soudack recommended expanding these extremes further qualify signals. A move above 90 is truly overbought and a move below 10 is truly oversold. Moves above 90 and below 10 are rare occurrences that suggest a price move is unsustainable.
Commodity Channel Index
1. CCI = (Typical Price - 20 - period SMA of TP) / (0.015 XMean Deviation)
2. Typical Price (TP) = (High + Low + Close) / 3
3. Constant = 0.015
Developed by Donald Lambert and featured in Commodities magazine in 1980, the Commodity Channel Index (CCI) is a versatile indicator that can use to identify a new trend or warn of extreme conditions. Lambert originally developed CCI to identify cyclical turns in commodities, but the indicator can successfully applied to indices, ETFs and other securities. In general, CCI measures the current price level relative to average price level over a given period. CCI is relatively high when prices are far above their average. CCI is relatively low when prices are far below their average. In this manner, CCI can use to
Ghobadi M., Ghobadi M., Heshmatpour M.
identify overbought and oversold levels. As noted above, the majority of CCI movement occurs between - 100 and +100. A move that exceeds this range shows unusual strength or weakness that can foreshadow an extended move. Think of these levels as bullish or bearish filters. Technically, CCI favors the bulls when positive and the bears when negative. However, using a simple zero line crossovers can result in many whipsaws. Although entry points will lag more, requiring a move above +100 for a bullish signal and a move below - 100 for a bearish signal reduces whipsaws.
Data
This paper applies this process to 10 listed Dow-johns companies for evaluation, namely, Yahoo, Face-book, Google, Apple, Microsoft, General Electric, Coca Cola, Cisco, Boeing and Bank of America. Data include 13200 daily prices "between (2008) to (2013)".
Step 3: Determine the ideal (A*) and negative ideal ( A ) solutions.
A* = {(maxvi} | j G Cb), (minvi} | j G C )} =
= {v*| j = 1,2,..., m} 2)
j
A = {(>mvj 1 j G CbX (maxvj 1 j G C )} =
= {v,. | j = 1,2,...,m} (3)
Step 4: Calculate the separation measures using the m-dimensional Euclidean distance. The separation measures of each alternative from the positive ideal solution
and the negative ideal solution, respectively, are as follows:
Si*-,Ev -v*)2 ,J = 1,2,---,m
J=1
s-^j(vj-vj)2 ,j = 1,2,- =
(4)
(5)
Research Methodology
1. T-Test: It can use to determine if two sets of data are significantly different from each other, and is most commonly apply when the test statistic would follow a normal distribution if the value of a scaling term in the test statistic known.
t
У - У2
ll
— +-
V nl
2 Z
Step 5: Calculate the relative closeness to the ideal solution. The relative closeness of the alternative Ai with respect to A* is defined as follows:
RC =—^, i = 1,2,...,m ¡ S + S.
I l
(б)
Step 6: Rank the preference order (see Table 1).
Results and discussion
2. TOPSIS: Topsis based on the concept that the chosen alternative should have the shortest geometric distance from the positive ideal solution and the longest geometric distance from the negative ideal solution
Step 1: Calculate the normalized decision matrix. The normalized value r.,. is calculated as follows:
m &
i = 1, 2, ..., m and j = 1, 2,
Step 2: Calculate the weighted normalized decision matrix. The weighted normalized value v is calculated as
V
follows:
v.. = r.. x w, i = 1, 2, ..., m and j = 1, 2, ..., n, (1)
V V J
where wV is the weight of the Vth criterion or attribute and
s
J=l
w. = l.
J
The multi-criteria decision-making method is a common replication applied in Operations Management, which is then modified to be applied in Financial Management scope. The first research in Financial Management done by Tarmizi (2006) in Indonesia. Therefore, this method is relatively new in Financial Management, especially for researches that are conducted in Indonesia. The modification is done by changing the criteria used in operations field into financial ratio criteria.
The application of statistical factors in this multi-criteria decision-making method is still considered trial. Thus, there has been no established financial criterion applied for this method until nowadays. Simultaneous research can be expected to develop some useful criteria, which can be a fundamental in stock selection method, as well as a new method in the Financial Management scope.
Technical analysis indicators can show the profitable trading prices.
1. STO with a mean of (0.059)
2. RSI with a mean of (0.055)
m
2
s
y
2
r
x
n
iJ
iJ
iJ
Ghobadi M., Ghobadi M., Heshmatpour M.
Table 1
ST.D NT NF MIN MAX MEAN
RSI 0.0045 357 142 -0.042 0.797 0.55
CCI 0.0041 400 306 -0.033 1.19 0.52
MFI 0.0057 277 159 -0.031 0.854 0.46
SMA 0.0048 320 135 -0.036 0.601 0.47
MACD 0.0047 223 120 -0.031 0.549 0.48
STO 0.0039 418 203 -0.029 1.28 0.59
MIN MAX MIN MAX MAX MAX
WEIGHTS 1 1 1 1 1 1
IDEAL 0.0039 418 120 -0.029 1.28 0.59
THE WORST 0.0057 223 306 -0.042 0.549 0.46
C1 C2 C3 C4 C2 C3
RSI 0.0012 -134 164 0 -0.248 -0.09
CCI 0.0016 -177 0 -0.009 -0.641 -0.06
MFI 0 -54 147 -0.011 -0.305 0
SMA 0.0009 -97 135 -0.006 -0.052 -0.01
MACD 0.001 0 186 -0.011 0 -0.02
STO 0.0018 -195 103 -0.013 -0.731 -0.13
NORMA 0.0030083 315.65012 334.56688 0.0229783 1.0499881 0.1705872
NORMED MATRIX
C1 C2 C3 C4 C2 C3
RSI 0.3988935 -0.4245207 0.490186 0 -0.2361932 -0.5275893
CCI 0.531858 -0.5607474 0 -0.3916747 -0.6104831 -0.3517262
MFI 0 -0.1710755 0.439374 -0.4787136 -0.2904795 0
SMA 0.2991701 -0.3073023 0.4035068 -0.2611165 -0.0495244 -0.058621
MACD 0.3324112 0 0.5559426 -0.4787136 0 -0.1172421
STO 0.5983402 -0.6177726 0.3078607 -0.5657524 -0.6961984 -0.7620735
WEIGHTED NORMED MATRIX
C1 C2 C3 C4 C2 C3 CI RESULT -RANK
RSI 0.39889 -0.42452 0.49018 0 -0.236193 -0.5275893 0.6907586 2
CCI 0.531858 -0.56074 0 -0.391674 -0.610483 -0.3517262 0.3655402 5
MFI 0 -0.17107 0.43937 -0.478713 -0.290479 0 0 6
SMA 0.299170 -0.30730 0.40350 -0.261116 -0.049524 -0.058621 0.5420576 3
MACD 0.332411 0 0.55594 -0.478713 0 -0.1172421 1 1
STO 0.598340 -0.61777 0.30786 -0.565752 -0.696198 -0.7620735 0.5225867 4
IDEAL 0.598340 0 0.55594 0 0 0
THE WORST -0.61777 0 -0.565752 -0.696198 -0.7620735
Ghobadi M., Ghobadi M., Heshmatpour M
3. CCI with a mean of (0.052)
4. MACD with a mean of (0.048)
5. SMA with a mean of (0.047)
6. MFI with a mean of (0.046)
The results provided strong support for the technical strategies. According to confidence level, %95 all the technical analysis indicators used in this research can find profitable trading prices and all the returns are more than zero.
Returns of trading based on technical analysis strategy are more than risk free interest.
We have to compare the returns for all the filters of each stock and the return of "risk-free interest rate". The filter that generates the maximum return among all is calling the optimal filter for that stock. The summary of results for this sub-period is described that According to the results of the TOPSIS in order STO with a mean of (0.059), RSI with a mean of (0.055), CCI with mean of (0.052), MACD with a mean of (0.048), SMA with a mean of (0.047) and MFI with a mean of (0.046) stand after each others in the list.
Conclusion
In this paper, the aim is to rank Technical analysis indicators while considering advantages of some of the important ranking methods, existing in literature. As each of existing ranking methods have some major benefits that other do not have and the Fact that it is not possible to gather all these advantages in a united model, thus is seems significant to provide a new ranking method which considered all the good aspects of these models. In doing so, MCDM method is considered. As regards of the obtained ranking orders form deferent ranking models and a matrix of weights, corresponds to the deferent property of these methods, TOPSIS is accounted for in order to consider deferent aspects of these methods and a new Method introduced. The summary of results for this sub-period describe that in row, STO with a mean of (0.059), RSI with a mean of (0.055), CCI with mean of (0.052), MACD with a mean of (0.048), and MFI with a mean of (0.046) everyone has an abnormal return greater than the risk-free interest rate with a mean of (0.004). For further research on this subject, other aspect of MCDM technique can also be accounted for in order to obtain a new ranking order on basis of the existing ranking methods.
References:
1. Akbar rahimipoor(2012), A Comparative Study of Bankruptcy Prediction Models of Fulmer and Spring-ate in Accepted Companies in Tehran Stock Exchange,
Caspian Journal of Applied Sciences Research, 1(13), pp. 76-81.
2. Barry Eichengreen, Andrew K. Rose and Charles Wyplosz. Contagious currency crisis. NBER Working Paper No. 5681, July 1996.
3. Benjamin Graham (1934), Security Analysis: Sixth Edition, Foreword by Warren Buffett (Security Analysis Prior Editions), McGraw-Hill.
4. Chan, F. T. S. & Chan, H. K., 2005, Design of a PCB plant with expert system and simulation approach. Expert Systems with Applications, 28, 409-423.
5. Chen, Y.W. and Ten, G.H. (2001), "Using fuzzy integral for evaluating subjectively perceived travel cost in a traffic assignment model", European Journal of Operation Research, 130(3): 653-664.
6. Chenard, J. S. & Chu, C. Y., 2005, Design methodology for wireless nodes with printed antennas, The 42nd Design Automation Conference.
7. Chiou, H.K. and Tzeng, G.H. (2002), "Fuzzy multicrite-ria decision making approach to analysis and evaluation of green engineering for industry", Environmental Management, 30(6): 816-830.
8. Cullen, A., 2002, Comparison of measured and predicted environmental PCB concentrations using simple compart-mental models, Environmental Science & Technology, 36, 2033-2038.
9. Eccles, Audley Delory (2003), Stock selection method: Fundamental analysis departure in the United States stock market compared and contrasted a dissertation of Ph.D. in finance. Nova Southeastern University. Economics 78, 187-202.
10. Elleuch, J. L. Trabelsi, (2009) "Fundamental Analysis Strategy and the Prediction of Stock Returns", International Research Journal of Finance and Economics, ISSN, 1450-2887, Issue30.
11. Ghobadi, Mohsen. 2014. "Profitability of Technical Analysis Indicators to Earn Abnormal Returns in International Exchange Markets." Paper presented at International Conference on Economic, Accounting, Management and Social Science, Szczecin, Poland, December 11.
12. Ghobadi, Mohsen. 2014. "Profitability of Technical Analysis Strategy to Earn Abnormal Returns in Tehran Stock Exchange (2007-2013)." M.A Thesis, Islamic Azad University, Dehaghan Branch, Iran.
13. Giachetti, R. E. & Alvi, M. I., 2001, an object-oriented information model for manufacturability analysis of printed circuit board fabrication. Computers in Industry, 45, 177-196.
14. Grunow, M., Günther, H. O. & Fohrenbach, A., 2000, Simulation-based performance analysis and optimization of electronics assembly equipment, International Journal of Production Research, 38, 4247-4259.
15. Hammad Hassan Mirza(2013), Does Investor Protection Affect Managements' Decision to Pay Dividend?
Ghobadi M., Ghobadi M., Heshmatpour M.
Evidence from South Asia, Caspian Journal of Applied Sciences Research 2(11), pp. 26-38.
16. Jacqueline Grosh ed. The LGT guide to World Equity markets 1996. Euromoney Publications PLC and LGT Asset Management, London.
17. Liu, S. T. & Wang, R. T., 2009, Efficiency measures of PCB manufacturing firms using relational two-stage data envelopment analysis, Expert Systems with Applications, 36, 4935-4939.
18. Lütkepohl, H. and Kratzig, M., 2004. Applied Time Series Econometrics (Themes in Modern Econometrics). Cambridge University Press.
19. Mark Andrew Lim (2014), a Handbook of Technical Analysis: The Practitioner's Comprehensive Guide to Technical Analysis, Wiley. Robert Aderhold, Christine Cumming, and Alison Harwood. International linkages among equities markets and the October 1987 market break. FRBNY Quarterly Review/ Summer 1988.
20. Opricovic, S. and Tzeng, G.H. (2003), "Defuzzification within a multicriteria decision model", International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 11(5):
21. René Carmona (2013), Statistical Analysis of Financial Data, Springer Texts in Statistics.
22. Rossetti, M. D. & Stanford, K. J. A., 2003, Group sequencing a PCB assembly system via an expected sequence
dependent setup heuristic. Computers and Industrial Engineering, 45, 231-254.
23. Ryan, A. & Lewis, H., 2007, manufacturing an environmentally friendly PCB using existing industrial processes and equipment. Robotics and Computer-Integrated Manufacturing, 23, 720-726.
24. Tadas Viskanta, Abnormal Returns (2012), Winning Strategies from the Frontlines of the Investment Blogosphere, the McGraw-Hill companies.
25. Tsai, C. Y. & Chiu, C. C., 2007, a case-based reasoning system for PCB principal process parameter identification. Expert Systems with Applications, 32, 1183-1193.
26. Tsai, C. Y., Chiu, C. C. & Chen, J. S., 2005, a case-based reasoning system for PCB defect prediction. Expert Systems with Applications, 28, 813-822.
27. Wang, R. T., Ho, C. T. and Oh, K., 2010, Measuring production and marketing efficiency using grey relation analysis and data envelopment analysis, International Journal of Production Research, 48(1), 183-199.
28. Watada, J. (2001), "Fuzzy portfolio model for decision making in investment", Dynamical Aspects in Fuzzy Decision Making, 141-162.
29. Yu, S., Kim, D. & Park, S., 2005, Integer programming approach to the printed circuit board grouping problem, International Journal of Production Research, 43, 16671684.
