KR20170011780A - Population statistic simulation method - Google Patents

Abstract

The present invention provides a population statistic simulation method including the steps of: generating the spread of initial population by using population statistical data; generating personal models as to each person of the initial population corresponding to the spread of the initial population generated by using the population statistical data, and generating an environment model through national index statistical data; forming a virtual society which comprises a group of multiple persons having the personal models and the environment model, and where behavior and social interaction among the personal models and the environment model are done; and outputting simulation result on the basis of the virtual society, and predicting population statistics. Therefore, it is possible to predict more integrative and reliable population statistics, by integrally considering a series of events in accordance with a life cycle including educational transition in accordance with economic situation, and a life cycle in accordance with birth and death by predicting the future (statistics) through a virtual society.

Description

[0002] Population statistic simulation method [

The present invention relates to a demographic simulation method, and more particularly, to a demographic simulation method for predicting population statistics by taking into consideration population birth, death, advancement, and employment.

Recently, we are studying various measures to solve population problems caused by low fertility and aging problems. This population problem is the basis of almost all national policies of a national society, It can be said that it affects.

Therefore, demographic simulations are being developed to analyze the population problem and to analyze the nature of social phenomena that affect demographics.

However, the above-mentioned conventional demographic simulation has a problem in that it is difficult to obtain various analysis results because it is constituted by analyzing individual births, deaths, and employment rates individually and obtaining statistical results therefrom.

Korean Patent Publication No. 10-2011-0102125

It is an object of the present invention to provide a demographic statistical simulation method that predicts demographics by taking into consideration the entrance examination, the death, the advancement, the employment, the entrance examination and the employment according to the life cycle of the individual and the economic situation.

The present invention relates to a demographic simulation method executed by a demographic forecasting simulation server, comprising the steps of generating an initial population distribution using demographic data, corresponding to the distribution of the initial population generated using the demographic data Generating individual models for each individual of the initial population and generating an environmental model through national indicator statistics, a plurality of groups of individuals having the individual models and the environment model, A step of forming a virtual society in which behavior and social interaction between the environment model and the environment model is performed, and outputting a simulation result based on the virtual society and predicting demographics.

The demographic simulation method according to the present invention provides the following effects.

First, by implementing a personal model based on demographics, a virtual society is constructed, and the future (statistics) is predicted, and a life cycle including a life cycle due to birth, death, It is possible to obtain a comprehensive and comprehensive analysis result rather than a more individual and fragmented analysis result.

Secondly, the personal model is created by taking into account the integrated information such as birth, death, advancement, employment, and income including the life cycle of the individual, and after implementing the virtual society based on this, Can be performed.

Third, a personal model is implemented based on comprehensive intellectual knowledge (Knowledge in multi-disciplinary), and a virtual society is constructed on the basis of it. .

1 is a flowchart illustrating a procedure of a demographic simulation method according to an embodiment of the present invention.
2 is an exemplary diagram illustrating an initial population generation in the demographic simulation method of FIG.
3 is an exemplary diagram illustrating generation of a private model in the demographic simulation method of FIG.
4 is an exemplary diagram illustrating a statistical based behavior definition in the demographic simulation method of FIG.
FIG. 5 is an exemplary diagram illustrating a rule-based behavior definition in the demographic simulation method of FIG. 1. FIG.
FIG. 6 is a diagram showing statistics-based action definitions and rule-based action definitions in the demographic simulation method of FIG. 1; FIG.
FIG. 7 is a diagram showing a model for predicting job-based population movement in the demographic simulation method of FIG. 1; FIG.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The demographic simulation method according to an embodiment of the present invention is implemented by a population prediction simulation server (model), and implements the personal model and the environment model experienced by the individual on the basis of the statistics in a model of the population prediction simulation server, Based on the models and the environmental model, a demographic simulation is performed after forming a virtual society.

Referring to FIG. 1, an initial population distribution is created using statistical data (S10). Here, the above statistical data use the actual demographical data of the National Statistical Office and the national statistical data, and the demographic data includes various contents such as the population, sex, residence, income, marital status, The statistical data includes various contents such as regional economic characteristics and environmental characteristics. In this way, information for demographic simulations is utilized with real demographics and national indicator statistics, resulting in more integrated and more reliable forecasts.

On the other hand, the distribution of the initial population is generated by combining the distribution of the population by region, sex, age, marital status, income and economic activity status using actual statistical data for the population for a specific year, To generate a distribution of

Referring to FIG. 2, the initial population generation is generated using actual demographic data. As shown in FIG. 2, by utilizing the actual population distribution according to the 2000 population census (regional population) Apply the distribution. In relation to income, income is applied to the population (income accounts, employment rates existed since 2000) through the 2000 income charts of regions, the 2000 population estimates, and the employment rates by province in 2000.

In FIG. 2, the corrected employment rate is the employment rate × (population 15 years) / (population 15 to 60 years), and the adjusted salary is the total disposable income / corrected employment rate × (population 15 to 60 years old).

As described above, when demographic data is acquired to generate an initial population, individual models and environmental models are created for each individual in the initial society through the demographic data and the national indicator statistical data (S20). In the personal model and environment model generation step, a step of setting a personal model behavior and generating corresponding to the distribution of the initial population, and a step of setting an initial value and a time-dependent value of the environment model based on actual statistics, . Herein, the step of setting the personal model action includes defining an action that can be simulated by a human being with respect to the personal model, including death, marriage, birth, employment, etc., Based behavior setting and statistical based action setting. That is, the personal model and the environment model generation are classified according to the initial population distribution, the personal model having the residence, the sex, the age, the human relationship, the marital status, the education level, the income, An environmental model is created by invoking an indicator that affects the individual model.

Referring to FIG. 3 showing the generated personal model, the personal model includes personal items including a residence, sex, age, family relationship, marital status, and education level based on the demographic data Respectively. In addition, the personal model generates economic items including employment and non-employment status, income level, and ownership in consideration of the employment rate and the average income of the demographic data.

As described above, when each item of the personal model is created, the behavior of the personal model and the environment model is defined. First, the personal model and the environmental model define statistics-based behavior based on the demographic data and the national indicator statistical data.

Based on the statistics-based behavior described above, the statistical-based action includes a fertility rate according to an individual's marital status, a marital status change rate, a mortality rate, and a migration, Step includes a method of using conditional probability in actual statistics as shown in Equation 1 and a method of obtaining and applying a regression model of a logistic function form by statistical analysis of a panel data set as shown in Equation 2. [

&Quot; (1) "

The rate of change of state = (the number of people who have changed state after a certain period in the cohort) / (the number of people in a particular cohort)

&Quot; (2) "

State change rate = 1 / (1 + exp (-x))

x = intercept value + Σ ((related variable) * (coefficient for the corresponding variable))

On the other hand, the statistical based action can be defined as a transition probability calculation method such as a mortality rate or a marital state transition probability.

Referring to FIG. 4, regarding the mortality rate, the mortality rate of the age can be defined as Equation 3. Referring to FIG. 4, the mortality rate of the 5 year old unmarried male is 36/99177, which is 3.63% do.

&Quot; (3) "

Mortality rate for that age = (number of deaths for that age) / (number of survivors for that age)

Next, with respect to the marital state transition probability, the marital state change rate of the age can be defined as shown in Equation (4). Referring to FIG. 4, the probability of the first marriage of a 5 year old unmarried man is 0/99177 0%. Here, the transition rate of the marriage state of the corresponding age does not correspond to the transition probability of the widowed state.

&Quot; (4) "

Rate of change in marital status of the age = (number of people of that age) / (number of survivors in the marital status of the age)

The definition of the statistical-based behavior (behavior) is a method of obtaining a regression model of a logistic function function by analyzing a statistical correlation with respect to a panel data set and the like as shown in Equation (5). If this is applied to the fertility rate, Equation 6 is obtained.

Equation (5)

The probability that a model will have a particular event = 1 / (1 + exp (-x))

x = intercept value + Σ ((related variable) * (coefficient for the corresponding variable))

&Quot; (6) "

Fertility rate = 1 / (1 + exp (-x))

Cfb * C_fb * C_fb * C_fb * C_lb + Fb * C_fb *

Here, x is the parameter, Intercept is the intercept, Age is the age, Mig is the immigration status, Coh is the cohabitation status, Edu is the education level, Lab is the economic activity status, lastBirth is the last period from the last birth, , C_age is the coefficient for the age, C_mig is the coefficient for the immigration status, C_coh is the coefficient for the dwelling state, C_edu is the coefficient for education level, C_lab is the coefficient for economic activity, C_lbr: And C_fba: means the coefficient for the age at first birth.

In the above description, the statistics-based behavior of the individual model is defined through the above-described mortality rate, marital state change rate, and fertility rate, but it is needless to say that it can be defined as various items according to the demographic statistics to be predicted in one embodiment . In addition, if the above two methods are utilized, the statistics can be extracted and applied through a time series statistical analysis model such as the known ARIMA, if they are not provided in the statistical data for a specific period.

The personal model defines a rule-based behavior for the demographics based on factors that influence the demographics of the country in the virtual society, in addition to the statistical-based behavior.

Specifically, the definition of the rule-based behavior is a setting for an action that is influenced by another individual or country state, including an individual's mate-matching and migration region selection, and as a simple example of this, In the initial pop generation, at the age of twenty, I am leaving at 20 years old, mate-matching, marriage with a bride moves a groom), after setting the child close to the parent after birth, the child can be set to move to age 20 again.

 Referring to FIG. 5, the subject of the fertility module is a woman who is between 15 and 50 years of age, and can set married and unmarried, respectively. The mate matching rule is defined as follows: 2. Reason 3. It can be set according to the state of Find Mate, and the rate of change of marriage and mortality can be defined through the statistical-based behavior setting method according to the marital life table.

FIG. 6 is a diagram of the statistical-based behavior and rule-based behavior. Referring to FIG. 6, when the marital status change is considered, the statistical-based marital status change is the marriage rate of the virtual population And the rule-based marital status change is determined by whether the virtual population is married according to the surrounding state at the present time point.

On the other hand, the environmental model is created to simulate a policy, a local characteristic, an environment, and the like that are not represented as a personal model, including an index that can affect an individual and simulate the environment. The environmental model includes regional economic characteristics, environmental characteristics, and educational characteristics that are generated by using national statistical data.

In other words, as described above, the environmental model can be set as a variable that changes the population, such as a marital rate, a birth rate, an employment rate, an economic situation, etc. as a result, , The living environment, the price of the residence, the job, and so on.

Here, the environment model may have a statistically extracted value as it is, like the personal model, or may have different values according to time through a rule of interacting with an individual.

After constructing the statistics-based society, the personal model and the environment model are created by defining the statistical-based action and the rule-based action according to the respective variables, and then a virtual society is formed based on the personal models and the environment model (S30) . Here, the virtual society may be defined as a space consisting of a plurality of groups of individuals having the individual models and the environment model, and a behavior and social interaction between the individual models and the environment model.

When the above-mentioned virtual society is formed, a population simulation model including birth, death, and population movement is set up based on the virtual society, and a life cycle (age) (By age, sex, marital status, and economic activity status) according to the required level (individual, city, district, province, country) And outputs a simulation result and predicts demographics (S40).

Referring to FIG. 7 as an example of such a demographic forecast, FIG. 7 shows a simulation for predicting job-based population migration. First, the initial population is created as economically active population and the non-economy population is created. And statistical based action and rule based action are defined for each economically active and inactive population. As a result, the economically active population is in the range of 20 to 60 years old. In the case of non-economically active population, it is considered that the models of 20 years old or older can move, and they are classified into the employed and unemployed. And the unemployed considers the movement according to the employment position. The non-economically active population is defined as the population moving over 60 years of age in the economically active population. The age group can be defined as 20 years old, 60 years old or older, Furthermore, the shift from the non-economically active population to the economically active population can be continued to be made at a certain percentage of the working-class population, taking into account the movement due to the above-mentioned fleet (workplace). In addition, age induced migration can be advanced according to the vacancy in the environmental model.

As described above, the demographic simulation method implements a personal model in a population prediction simulation server based on comprehensive intellectual information and knowledge based on actual demographic data, Demographic simulations are conducted in society. These simulated values can be used to analyze storage and input / output, share input / output through feedback, and display higher predictive power.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

Claims (8)

1. A demographic simulation method executed by a population prediction simulation server,
Generating an initial population distribution using demographic data and national indicator statistics;
Generating individual models for each individual of the initial population corresponding to the distribution of the initial population generated using the demographic data, and generating an environmental model through the national indicator statistical data;
Forming a virtual society consisting of a plurality of groups of individuals having the individual models and the environment model, wherein behavior and social interaction between the individual models and the environment model are performed; And
Outputting simulation results based on the virtual community, and predicting demographics. The method according to claim 1,
The generation of the distribution of the initial population,
A demographic simulation method including the population distribution by year, the population distribution by region, the population distribution by age, and the population distribution by sex based on the demographic data. The method according to claim 1,
In the personal model,
A personal item including a residence generated based on the demographic data, sex, age, family relationship, marital status, education level,
A demographic simulation method that includes employment and non-employment status generated by considering the employment rate and average income of the demographic data, and economic items including income levels. The method according to claim 1,
Wherein the generating the personal model and the environment model comprises:
Defining statistical based actions based on the demographic data and the national indicator statistical data;
And defining a rule-based behavior for the demographic in the virtual community. The method of claim 4,
The step of defining the statistical based behavior comprises:
A demographic simulation method including a conditional probability calculated by Equation (1) and a state change rate through a regression model calculated by Equation (2).
(1)
The rate of change of state = (the number of people who have changed state after a certain period in the cohort) / (the number of people in a particular cohort)
(2)
State change rate = 1 / (1 + exp (-x))
x = intercept value + Σ ((related variable) * (coefficient for the corresponding variable)) The method of claim 4,
The statistical-
A marital status change rate, a mortality rate and a migration status according to the marital status of the individual,
The rule-
A demographic simulation method comprising the individual's mate matching rule and the migration region selection. The method according to any one of claims 1 to 6,
The demographic forecast predicts job-based population movements,
The occupation-based population movement prediction includes:
And calculating and predicting population movements according to the life cycle and the economic situation for each individual having the personal model in the virtual society. The method according to claim 1,
In the environmental model,
It is composed of economic characteristics of the region, environmental characteristics for residence and living, and educational characteristics, which are generated by using the national indicator statistical data, and the marital rate, the birth rate, the employment rate, the economy Set variables that change the population, including the situation,
Wherein the personal model comprises at least one of education level, residential environment, residence price, and job.

Images