This paper unveil the role of computer in financial decision making, it involves decisions concerning financial related issues such as determining the proper amount of funds to employ in a firm, capital expenditure analysis, forecast stock market analysis. To prognosticate the stock market, a machine has been designed by petr dostal i.e., stock market decision making machine to improve the accuracy in prediction using fundamental analysis, psychological analysis and technical analysis with simulation and other methods. They are Fuzzy logic, genetic algorithm, dynamic models, hybrid models, different methods, approaches, techniques are required and technical analysis is represented by regression and chaos analysis.
Keywords: decision making, stock market, fundamental analysis, psychological analysis, technical analysis, methods, models, approaches.
The purpose of this paper is to discuss about decision making process, intervention of computer in decision making process, different types of decision making computer systems, benefit of computer in financial decision making and explain the role of computer in trade and stock market prediction and decision making. Prediction of stock market is made using stock market decision making machine designed by petr dostal, its goal is to predict the right moment to sell, buy, hold, currency to obtain profit. The important aspect of this task is always connected with prediction of uninterrupted time series.
Overall demeanor of stock market is influenced by political factors, economical, psychological issues so the prediction is complex in trade and stock markets. In the real world scenario, if we want to make an entry into or exit out of a given stock we have to contact any stock broker or brokerage house to make a suitable decision. The solution provided by stock broker or brokerage house may not be perfect it may vary accordingly to the circumstances in overall global market like political factors, current state of economy etc. This is the reason why prediction has been a very tricky task. In this paper we are also going to discuss how stock market decision making machine helps in prediction of trade and stock market using computing techniques, models, technical analysis etc.
Decision making, it is gaining knowledge of identifying and choosing alternatives based on values and druthers of the decision maker. Making a decision connote in many alternatives provided we have to choose the best way that has maximum rate of success and best tallies with our goals.
Concepts of Decision Making:
Information: This is the information about decision, alternative solutions, probability of success rate of each alternative etc. All the information is gathered from that data we have to choose a best alternative which best fits to the problem.
Alternatives: These are potentially able solutions we have to choose from.
Criteria: These are the attributes that each alternative must have.
Goals: What are the best alternatives we have to select from the given possible alternatives? We have to select the most accurately suitable alternative to achieve a goal and to achieve overall objective a goal.
Value: Value of the alternative i.e., cost and advantages, output.
Preferences: It deals with the psychology of decision maker, some prefer complex to normal and steady to fast it depends on individual choice. It deals with the personal opinion or preferences.
Decision Quality: This provides whether a decision is good or bad. It depends on the information provided, best alternative selected, preferences of decision maker.
We must consider one important aspect here, quality of decision is not related to its output. A good decision can have good or bad output in the same way bad decision may have good output. When it comes to the judging the quality of a decision made we have to consider few aspects. The decision made in the selection of alternative is important i.e., the alternative chosen must meet the goals, objectives, cost and effects must be taken into account.
The Process of Decision Making:
Gathering information and resources: This information comes from varies sources through different means. Reports generated by different computer systems assist management in decision making. Examples are sales and productivity report etc.
Understanding the various sides of issues: Advancement in the area of computing has made it possible to systems that attempt to enhance the performance of one or more human experts. We have to analyze various requirements needed to a firm. Applications such as Microsoft Excel can be arranged as an expert system to help in decision making.
Evaluating, Weighing and Comparing: We have to evaluate, weigh and compare the alternatives that can easily fit to solve the solution. In Financial decision making, Computer Systems help in manipulating, formatting and distributing the data. Several tools are provided so that user can easily understand the language and easily evaluate, weigh and compare data. Examples are Pie charts, Audit Command Language are useful.
Choose the Best Alternative: This is the main part of the task i.e., to select the suitable alternative for making decision. After evaluating alternatives grid analysis and decision trees analysis are the tools useful to choose the best alternative.
Make Decision or Implement Decision: With all of the effort and hard work in evaluating alternatives and deciding the best suitable way of decision making. Depending on the nature decision has to be made. In the financial sector, implementing of decision are also made through computer aided tools. Such as, Computer aided software engineering (CASE), Computer-aided recording tools (CART), Computer aided summarization tool (CAST).
Check Result: This is the final phase for process of decision making, it provides the outcome whether a decision is good or bad. It depends on the information provided, best alternative selected.
Different Types of Decision Making Computer Systems:
Decision Support Systems (DSS): Information Systems researchers and technologists have built and investigated Decision Support Systems (DSS) for approximately 40 years. Decision Support Systems (DSS) is a computerized information system that helps in business and organizational decision-making activities. DSS is an interactive software-based system intended to help decision makers compile useful information from raw data, documents, personal knowledge, and/or business models to identify and solve problems and make decisions. Typical information that a decision support application might gather and present would be accessing all current information assets, including legacy and relational data sources, data marts, revenue figures, economic calculations of a firm and comparing the sales achieved by the company.
Executive Information Systems (EIS): It is a type of management information system determined to simplify and support the information and decision-making needs of senior executives by providing easy access to meet goals of the organization. It is commonly considered as a specialized form of a DSS.
Executive information system allows information to collected and presented to the user without modifying or further processing. So that user can quickly see the data of his chosen department, enabling them to focus on decision making. It provides strong reporting and data mining efficacy which can provide all the data needed by executives. The two main aspects of an EIS system are integration and visualization. The new method of visualization is the Dashboard and Scorecard. Dashboard, it represents main or important data and organizational information in real time and integrated basis. The Scorecard is another one screen, which displays with measurement metrics. There are a number of ways to link decision making to organizational performance. In the view of decision maker these tools provide an excellent way of viewing data.
Expert Systems (ES): Expert Systems are computer systems which posses the knowledge by an experienced expert and imitate the expert and act as a consultant in a particular area.
Further we are going to discuss about the role of computer in financial management in the area of trade and stock market analysis. As we discussed earlier we cannot predict the stock market accurately it may vary frequently accordingly to the global, local issues, political factors, economy state. To predict future increment or decrement prices of shares, currency rate and commodities during the operation in the world market. The outcome from fundamental, psychological, technical analyses and other methods can be applied to get these data. The soft computing includes theory of fuzzy logic, neural network and genetic algorithms these methods help to provide better description of non linear processes that involve in business economy and finance. The time series analyses and prediction is described in this paper. This prediction includes fundamental analyses, psychological analyses, dynamic models, fuzzy logic models, neural networks models, genetic algorithms models, regression analyses, chaos analyses, hybrid systems, simulation and other methods. 
The model is presented in the following figure and the description is as follows:
- Information: can be received or gathered from print media, radio and internet and so on. Examples are news disclosed from firms, plans of firm, management, economic information such as Earnings per Share, Return of Equity, Price to Earnings Ratio, future earning potentials, dividends, income, debt, returns, income statement, balance sheet and so on. As well the values dealing with the history of data of prices might be obtained with various samplings, i.e. to obtain a time series. The time series can be processed by means of mathematics analyses with their advantages and disadvantages.
- Data mining: Data mining refers to extracting knowledge from large amount of data. Knowledge discovery steps consist of iterative sequence of the following steps: data cleaning is used to remove noise or irrelevant data, data integration is where multiple data sources may be combined, data selection is where data relevant to the analysis task are retrieved from the database, data transformation is where data are transformed or consolidated into forms appropriate for mining by performing summary or aggregation operations, pattern evaluation and knowledge presentation. 
- Fundamental Analysis: The information gathered is processed by fundamental analyses. E.g. the Delphi method.
- As it is input analysis the methods of determining whether the time series is fixed or settled or random could be used. Having proved that the time series is purely random, another search with the aim to determine the future development of price is tangential. We can use the Lyapunov exponent, which determines the range of predictability of time series. If the time series is not random and has a good predictability then it is suitable to continue in other analyses.
- Psychological Analysis: We can get analysis by Elliott's waves, which applies the knowledge of psychological analysis. It is possible to regulate the future trend of future price from the types of waves.
- Technical Analysis: It is possible to study the time series by means of regression analysis, dynamic models, fuzzy logic, neural networks, genetic algorithms, chaos analysis, hybrid models, wavelet analysis.
- Regression Analysis: It is possible to analyse time series using regression analyses . We can feign buying and selling signals in the past in so that we can obtain a view of the behaviour of the price depending on time and we can also estimate the future possible trends in price.
- Dynamic Models: It is possible to make a calculation in prediction of time series by means of the dynamic models represented by means of non seasonal and seasonal ARIMA (Autoregressive Integrated Moving Average) models and by various types of "smooth" models.
- Fuzzy Logic: We can also predict time series and calculate by means of fuzzy logic. The trend prediction of time series is calculated on the basis of fuzzy rules. The process of fuzzification, fuzzy inference and defuzzification are used.
- Neural Network: We can calculate prediction of time series by means of neural networks. The sigmoid and hyperbolic tangent are one of the best for prediction.
- Genetic Algorithm: Another possible way to calculate the trend identification of prediction of time series is by the means of genetic algorithms. The trend prediction is calculated on the basis of logical rules when the maximum profit is being searched by means of optimization. The various algorithms can be used for prediction of time series, for example parallel evolutionary ones.
- The chaos analysis represents the calculation and evaluation of Hurst and Lyapunov exponents it is same as D.
- Hybrid Systems: These systems can be created by the combination of fuzzy logic, neural networks, genetic algorithms and the theory of chaos. There can be various combinations where the neural - fuzzy- genetic - chaos one is the most complicated.
- It is possible to analyse time series in a post mortal way by wavelet analyses. We can use for decomposition of time series the wavelet function according Daubechies. The future values can be estimated by means of Kalman filter.
- Simulation: It is the process that explains the behaviour of stock market using differential equations. As stock market is a financial sector it can have negative and positive feedback.
- Other outputs can be based on the intuitive prediction or exotic analysis.
- Decision Making Process: Decision Making Process is very complicated and important task. If we get any conflicting outputs from these analyses, it is suitable to apply the fuzzy logic for the evaluation.
- Evaluations: It is important to calculate and compare the result actual values and with the predicted values.
- The signals for purchasing or selling can be obtained from different ways by means of e-business through internet etc.
Lotfi zadeh introduced the theory of Fuzzy Logic in his paper, Fuzzy Sets (1965). Fuzzy Logic is a method provides how to reduce complexity and explains the system complexity. Fuzzy models like traditional Expert and Decision Support System is based on the flow of input and output flow pattern.
The process of decision making is made using fuzzy logic in the following way. The outcome of fundamental analyses gives the information about economic data, balance sheets like earnings per share, profit per Earnings Ratio, future earning potentials like good, bad, average, very good, worst. The outcome of psychological analyses is the trend developed by the public behaviour on trade and stock market like steady, increase, decrease. The results of technical analyses will be processed from regression analyses, fuzzy logic, genetic algorithms, chaos analyses, neural networks, dynamic models. The result of simulation and various different methods indicate different trends in time series. The diagram represented below shows the fuzzy logic decision making process in that C,E,G,H,I,J,K,L,M,N,O,P as inputs Q is output.
The final result of all the analyses i.e., input for fuzzy logic which is executed by fuzzy rules. Fuzzy rules are A and B: The errors occurred by prediction from different analysis are used in fuzzy rules to set weights.
Fuzzy Logic Decision Making Process Diagram:
The working of Decision Making Machine is shown below: The outputs individual analyses and predictions are transferred to the attributes: High Positive, Middle Positive, Low Positive, Neutral, Low Negative, Middle Negative and High Negative.
The output result of technical analyses are executed in the form of inexplicit process from regression and chaos analyses and prediction of trends is made by dynamic models, fuzzy logic models, neural networks and genetic algorithms. Trends development is denoted as daily, weekly, monthly prediction of values such as price of share and currency profit. The outcome of psychological analyses will depend on public behaviour on trade and stock market. Fundamental analyses give the information about economic data, balance sheets like earnings per share, profit per Earnings Ratio and future earning potentials. The results of simulation and various different methods indicate different trends in time series.
The final result of all the analyses i.e., input for fuzzy logic are executed and associated to functions. Fuzzy rules are A and B: The errors occurred by prediction from different analysis are used in fuzzy rules to set weights. Fuzzy rules are effected by intensity of support. To examine and judge carefully errors came out of the analyses while predictions are used to set intensity of support. The output produced by fuzzy rules are measured from -100% to +100%. The scale provided on decision making machine help us to whether sell or purchase shares according to reading on machine. On whole individual models, methods, techniques, approaches, analyses applied or executed helps in decision making in trade and stock market.
In this paper we have discussed about the role of computer in financial management and with the help of computing how we are able to predict trade and stock market and also discussed about decision making process, intervention of computer in decision making process, different types of decision making computer systems and the role of computer in trade and stock market prediction and decision making. On the overall discussion made above gives us the benefit of computer systems in financial sector they are:
- Information produced from the computer systems is used in financial decision making.
- Computer helps us to make quick and fast track data analysis for decision making.
- Computers also aid to increases accuracy and trim down errors.
- Computers helps to control the decision making process.
- It unwraps the new way of approach to thinking about the problem space.
- Also helps to generate new proof in support of a decision making.
- Information provided computer systems is easily understood and provides skillfulness in avoiding wasted time and effort for decision makers.
- It adds consistency to the decision making process.
If we consider about decision making in trade and stock market using stock market decision making machine, it is helpful to predict nature of shares in present and future market trends, commodities in world market. Role of computing using Fundamental analyses, psychological analyses, simulations, models, Techniques, approaches e.t.c plays a major in this process.
Every year 85% of the people who started their investment in stock market doesn't get fruitful results almost they lose their money and few people quit the market in the first year itself. Stock market is like an ocean, if you swim well you will reach shore if not you will be lost.
In conclusion I would like to say, Computers and computing plays a major role in financial management and also in Trade and Stock Market. We had analyzed about stock market decision making machine designed by petr dostal which helps us to some extent to predict future trends of shares, currency, mutual funds etc beside these there is a complexity in that model i.e., random tuning and calculations has to be done in time series. My view on stock market prediction is, it is very tricky and lot of risk, hard work involved in trading. We can't estimate and foretell what could happen next minute, if we purchase shares of any company by the information or prediction by stock broker or by the machine. In the next minute or day we can lose our money if any incident happens or change of government policies. The best example is when WTC Towers were attacked and Recession in 2008 at that time they many people lost their money who invested in shares instead of good prediction and evaluation of market. Computer-Computing plays an important role in every sector in financial, trade and stock market but we can't foretell what will happen so if everything is fine in global market then predictions are correct and needful.
- Financial Management, I M Pandey.
- A Brief History of Decision Support Systems, D.J Power editor, www.dssresources.com
- Data Mining, Jiawel Han and Micheline Kamber.