Syzdykova E.Zh. – candidate of economic sciences

Kassenova D.I. - master of economic science

Koshmagambetova K.D. - master student

Academician Y.A.Buketov Karagandy State University

Financial prediction and it’s methods

In modern conditions managing decisions should be made only after a careful analysis of available information. For example, a bank or a board of directors of a corporation should decide investing money into a project only after careful calculations, coupled with forecasts of market conditions, determining return on investment andappraising the possible risks. Otherwise, it may surpass its competitors, who can better assess and predict the prospects for development.

To solve problems associated with data analysis in the presence of random effects, there is a powerful apparatus of Applied Statistics, part of which is a statistical method of forecasting. These methods allow the identification of patterns on the background of chance, to make informed predictions and assess the likelihood of their implementation.

Forecast is a scientifically based description of the possible states of the objects in the future, as well as alternative routes and timetables to achieve this state. The process of developing forecasts is called prediction (Greek Prognosis - foresight, prediction).

Forecasting must respond to:

• What is likely to expect in the future?

• How to change the conditions to achieve a specified, finite state projection of an object?

Forecasts that respond to the questions of the first type, called the search forecasts, the second type – regulatory forecasts. For example, the aim is to provide each family with a separate apartment with an improved layout. Normative forecasts show investment and the time within which this task would be accomplished.

Depending on the prediction object there areseparate forecasts for the scientific, technical, economic, social, political and military purposes, etc. However, this classification is conditional, since mainlybetween these forecaststhere are many forward and backward linkages. The need to determine final practical prediction purposes involves determining the desired type of prediction. Type of forecast is determined by two factors:

- Forecasting horizon and

 -Hierarchical level of the indicator topredict.

By the time horizon forecasts are divided into short (1-2 cycles ahead of time), medium (3-5 cycles) and long (more than 5 strokes ahead of time).

By the level of the indicator topredictthey are divided into macro-, meso- and micropredictions. Everything related to the prediction of parameters describing the activities of firms, companies and enterprises related to the micro level. Meso (regional and sectoral levels) and macropredictionsare used to describe the external environment.

Classification of economic forecasts is shown in Figure 1.

After studying Figure 1 we see that economic forecasts differ on the following criteria: the magnitude of the object prediction of lead time, for purposes of forecasting.

Construction of forecast and the related construction and experimental check (verification) probabilistic-statistical model is usually based on the simultaneous use of two types of information:

- A priori information about the nature and essence of the content of the phenomenon represented, usually in the form of some theoretical laws, restrictions, and hypothesis;

- Source of statistical data on the process and results of functioning of the phenomenon or system.

It should be emphasized that in reality, a businessman, head of the company may of coursesuccessfully run business and do not use methods of constructing mathematical models of forecasting. However, in the competitionusing these techniques provides a businessman and his business at times no less significant competitive advantages than conquesting of a certain market share or getting the best loan.

Figure 1. Classification of economic forecasts

Most practical interest of course representthe short-term and operational forecasts. Mathematical methods and models used in problems of stochastic analysis and forecasting in business can relate to the various branches of mathematics: a regression analysis, time series analysis, formation and evaluation of expert opinions, simulation, systems of simultaneous equations, discriminant analysis, logit and probit models, the unit of logical decision functions, the dispersion or covariance analysis, rank correlations and contingency tables, etc. However, they are all united by the fact that represent different approaches to solving the central problem of multivariate statistics and econometrics - the statistical study of dependency problems, which, again, is the basic problem of statistical analysis and forecasting in business. The scheme of choosing the forecasting mathematical tools is shown in Figure 2.

Figure 2. The scheme of choosing the forecasting mathematical tools

Thus, the prediction is a multistage process involving goal setting, information, processing, assessment and analysis, identification of prospects and the probability of the forecast. At all stages of the forecasting a close alignment with the objectives and tasksmust be ensured, i.e. with the strategy of economic development, while in some cases on the basis of our stated goals are projected path and its achievements, in other cases, the forecast is carried out to determine the achievable, responsive to the needs of society goals.

References:

1. Aivazian S.A., Mkhitaryan V.S. Applied statistics and the foundation of econometrics. - Moscow: UNITY, 2008.

2. Hanke J. and Reitsch A. Business Forecasting. - N.Y.: Prentice-Hall, Englewood Cliffs, 2005.