Basic principles of market forecasting

Prophets for the future are not just what human beings yearn for, so predictions have long existed, including the myth that "knowing 500 years later will know 500 years ago", so predictions have been reduced to a disgraceful predecessor, called "divination".

If an enterprise can achieve a certain degree of foresight, the benefits to its operation are of course self-evident. Of course, it is impossible to make a complete prediction, otherwise everyone is a millionaire and every enterprise is thriving.

Although enterprises are uncertain about the future, the progress of human understanding and thinking has made people discover the importance of law. The ancients have long said that "Tao" is the law. With the accumulation of historical experience and the progress of science and technology, the ability of human beings to understand nature has been greatly enhanced. As an enterprise, discovering, understanding and applying "laws" (including market, customers, technology and enterprise development) will definitely increase the chances of winning.

：

Basic principles of prediction

In the simplest terms: it is the following mode.

Law, trend, logic, experience and essence are the ability and means to analyze problems.

Four principles of prediction

Prediction itself depends on mathematics, statistics and other methodologies and advanced means. Let's not talk about technology and methods. For managers of enterprises, the first concern may be how to form an effective way of thinking. The following principles may be instructive:

1, correlation principle: based on the thinking height of "classification", we pay attention to the correlation between things (categories) and infer the changing trend of another thing when we know (or assume) that something has changed.

The most typical correlations are positive correlation and negative correlation. In thinking, it is not entirely data correlation, but more "qualitative".

(1) Positive correlation is the "promotion" between things,

(2) Negative correlation refers to the mutual "restriction" between things, and the development of one thing leads to the restriction of another.

2. Inertia principle. Everything has a certain inertia, that is, it keeps its original trend and state at a certain time and under certain conditions, which is also the theoretical basis of most traditional forecasting methods.

3. The principle of analogy. This principle is also based on the thinking height of "classification" and pays attention to the correlation between things.

(1) From small to large-infer the general trend of things from a certain phenomenon.

(2) Inferring essence from exterior to interior-from superficial phenomena;

(3) The introduction of foreign advanced management and technology can also be explained by this idea.

(4) From the past and the present to the future-

(5) From far to near-for example, foreign products, technologies, management models, marketing experience and methods, because they may be advanced, represent the advanced direction, which may be "the way to go tomorrow".

(6) Bottom-up-infer the overall situation from the typical part.

(7) Top-down-subdividing from the whole, so as to know and infer a part.

4. Probabilistic reasoning principle. We can't fully grasp the future, but according to experience and history, we can often roughly estimate the probability of a thing happening and take corresponding measures according to this possibility.

To do a good job in forecasting, we must grasp the four basic elements of forecasting:

1, information Information is the representation and reflection of the characteristics and changes of objective things, which exists in various carriers and is the main working object, working basis and result reflection of prediction.

2. method. Methods refer to various means used for qualitative and quantitative analysis in the process of forecasting. According to different standards, forecasting methods can be divided into different categories. According to the attributes of the prediction results, it can be divided into qualitative prediction and quantitative prediction, and according to the length of the prediction, it can be divided into long-term prediction, medium-term prediction and short-term prediction. According to the method itself, it can be divided into many categories, and the most basic ones are model prediction and non-model prediction.

3. analysis. Analysis is a thinking research activity based on relevant theories. After drawing a prediction conclusion according to the prediction method, we must analyze two aspects: first, whether the prediction result meets the conditions of economic theory and statistical analysis needs theoretical analysis; The second is to analyze the accuracy of prediction error in practice and evaluate the reliability of prediction results.

4. Judges. It is necessary to judge whether the forecast results are adopted or whether the forecast results are revised according to the relevant economic and market trends, as well as the choice of judgment information and forecasting methods. Judgment is an important factor in forecasting technology. Edit this paragraph | The basic steps of the market forecast of return rate should follow certain procedures and steps to make the work orderly and co-ordinate. The process of market forecasting roughly includes the following steps:

1, determine the forecast target.

Defining the purpose is the first step in market forecasting, because different purposes will lead to different forecasting contents and items, required data and methods. A clear forecast goal is to draw up forecast items, make forecast work plans, prepare budgets, allocate forces and organize their implementation according to the problems existing in business activities, so as to ensure that the market forecast work is carried out in a planned and rhythmic manner.

Step 2 collect information

Market forecast must have enough information. With sufficient information, we can provide reliable basis for market forecast to analyze and judge. Under the guidance of market forecasting plan, investigating and collecting relevant information is an important part of market forecasting and the basic work of forecasting.

3. Choose a forecasting method

According to the target of prediction and the applicable conditions and performance of various prediction methods, the appropriate prediction method is selected. Sometimes multiple prediction methods can be used to predict the same goal. Whether the prediction method is appropriate or not will directly affect the accuracy and reliability of prediction. The core of applying prediction method is to establish a model to describe and summarize the characteristics of the research object and its changing law, and calculate or process it according to the model to get the prediction result.

4, forecast analysis and correction

Analysis and judgment is a comprehensive analysis of the data collected by investigation. Through judgment and reasoning, perceptual knowledge is raised to rational knowledge, and the phenomenon of things goes deep into the essence of things, thus predicting the future development and change trend of the market. On the basis of analysis and evaluation, it is usually necessary to evaluate and correct the original prediction results according to the latest information.

5. Prepare forecast report

The forecast report shall summarize the main activities of the forecast research, including the forecast target, the analysis conclusion of the forecast object and related factors, the main data and materials, the selection of the forecast method and the establishment of the model, as well as the evaluation, analysis and correction of the forecast conclusion. Edit this paragraph | Go back to the top market forecast. The content of market forecast is very extensive and rich. From macro to micro, the two are interrelated and complement each other. Specifically, it mainly includes the following contents: 1. Forecast market capacity and changes. Market commodity capacity refers to the total demand with a certain monetary payment ability. Market capacity and its change forecast can be divided into production material market forecast and consumption material market forecast. Forecasting the market capacity of means of production is to predict the demand structure, quantity and changing trend of industrial products by studying the development direction and focus of the national economy, comprehensively analyzing the adjustment of production technology and product structure during the forecasting period. The market capacity forecast of consumption data mainly focuses on the following three aspects:

(1) Consumer purchasing power forecast. To predict the purchasing power of consumers, we must make two predictions: one is the population and its changes. The population and its development speed largely determine the consumption level of consumers. Second, the forecast of consumers' monetary income and expenditure.

(2) Forecast purchasing power investment. The level of consumer income determines the consumption structure, that is, the ratio of commodity consumption expenditure to non-commodity consumption expenditure in consumer living expenditure. The law of consumption structure is that the higher the income level, the higher the non-commodity consumption expenditure, such as entertainment, recreation and labor expenditure, and the proportion of food expenditure in commodity expenditure is greatly reduced. In addition, we should fully consider the influence of consumer psychology on purchasing power investment.

(3) Forecast the change of commodity demand and its development trend. According to the total purchasing power and purchasing power input of consumers, the quantity, color, variety, specification and quality of various commodity demands are predicted.

2. Forecast the change of market price. The input price and product sales price in enterprise production are directly related to the profit level of enterprises. In commodity price forecasting, we should fully study the changes of labor productivity, production cost and profit, the development trend of market supply and demand, the changes of currency value and currency circulation, and the influence of national economic policies on commodity prices.

3. Forecast production development and its changing trend. The prediction of production development and its changing trend is the prediction of commodity supply and its changing trend in the market. Edit this paragraph | There are many ways to return to the top market forecasting method, including the following.

First, the time series prediction method

In market forecasting, we often encounter a series of economic indicators that change with time, such as the annual (quarterly) sales volume of a product of an enterprise, the income of consumers over the years, the statistical value of purchasing power growth and so on. These data arranged in time sequence are called time series. The method of forecasting with time series is called time series forecasting.

Second, the regression prediction method

1. The meaning of "regression". Regression is used to analyze and study the dependence between a variable (dependent variable) and one or several other variables (independent variables), and its purpose is to estimate or predict the overall mean of the dependent variable according to a set of known independent variable data values. In economic forecasting, people regard the predicted object (economic indicators) as the dependent variable, and those influencing factors closely related to the predicted object as the independent variables. According to the historical and present statistical data of the two, a regression model is established and used for forecasting after statistical test. Regression prediction includes univariate regression prediction with one independent variable and multivariate regression prediction with multiple independent variables. Only the univariate linear regression prediction method is discussed here.

2. Basic conditions of regression analysis. When using a set of known independent variable data to estimate and predict the value of the dependent variable, these two variables need to meet the following two conditions:

First, statistical correlation. Statistical correlation is an uncertain functional relationship, that is, the value of dependent variable (predictive variable) is obviously related to the value of one or more independent variables, but it cannot be accurately and uniquely determined, and all variables are random functional relationships. This correlation exists in a large number of economic phenomena. For example, the relationship between grain yield per mu y and fertilization amount x is obviously related, but there is no strict functional relationship. The yield per mu is not only related to the amount of fertilization, but also related to factors such as soil, rainfall and temperature, so the yield per mu y is random.

Second, causality. If one or several independent variables X changes, it will affect another variable Y according to a certain law, but the change of Y cannot affect X, that is, the change of X is the cause of the change of Y, not the other way around, that is, there is a causal relationship between X and Y, and the model reflecting the causal relationship is called regression model.

Another classification method of market forecast can generally be divided into qualitative forecast and quantitative forecast. For enterprise marketing managers, the enterprise forecasting methods that should be understood and mastered mainly include:

(1) qualitative prediction method

Qualitative forecasting method, also known as intuitive judgment method, is often used in market forecasting. Qualitative prediction mainly depends on the information, experience and comprehensive judgment ability mastered by forecasters to predict the future situation and development trend of the market. This prediction method is simple and easy, especially suitable for those problems that are difficult to obtain comprehensive data for statistical analysis. Therefore, qualitative forecasting methods are widely used in market forecasting. Qualitative forecasting methods include: expert meeting method, Delphi method, sales staff opinion collection method and customer demand intention survey method.

(2) quantitative prediction method

Quantitative forecasting is to use relatively complete historical data, mathematical models and measurement methods to predict future market demand. Quantitative forecasting is basically divided into two categories, one is time series model, and the other is causality model.

：

Market forecasting is a process of analyzing and inferring the future supply and demand trend, influencing factors and their changing rules of commodities in the market through analysis and research and using scientific forecasting techniques and methods.