Agent memory

As already mentioned in the General Model Layout, the agent behavior is depicted by stochastic functions, transforming memory variables of the agents or message contents into other messages or altered memory variables. But a major problem is that some of the variables needed to calculate the behavior (e.g. relationship status), are themselves dependent from other variables (e.g. age). So, where is our starting point?

Fixed, exogenous Variables
Let’s first determine some variables which we further deem to have no underlying stochastic process: ID, BIRTHDAY, SEX (male, female) and sexual orientation (SEXOR). (This is, of course, ignoring the possibility of intersexual persons as well as sex change. But all models are simplifications, so is ours.) These variables will not be changed during the simulation by interactions or agent behavior. As their values are given exogenously, let’s call them fixed, exogenous variables.

Flexible, exogenous Variables
If the agent is ALIVE, its AGE, the ID of its mother and father (MO, FA) and its probability to die (DEATHP), are variables which are also exogenously determined. But as the AGE increases on each birthday or the probability to die changes with increasing age, these variables are called flexible, exogenous variables. Therefore we already need some simple behavior rules for the agents to alter the variables’ values.

Flexible, endogenous Variables
The last category of variables is called flexible, endogenous variables as their values are set inside the model during the simulation as the results of different stochastic processes, i.e. agent behavior. The pregnancy-status (PREG) of an agent, the number of children (NOB), the ID and SEX of its partner (PID, PSEX) as well as the beginn of the current relationship (BCR), the number of past partnerships (NOP) or single sexual contacts (NSM), the usage of condoms or birth-control-pill (CONUSAGE, PILLUSAGE) and, finally, the state of syphilis, its duration and the infector-ID (SYPH, SYPHDUR, SYPHINFECTOR) are the result of complex, stochastic behavior rules of agents and interactions between agents.

Name Abbreviation Description
Fixed, exogenous variables
Identifier ID Unique identifier for each agent
Sex of agent SEX 0: female
1: male
Birthday BD Model iteration on which agent was born and AGE is incremented by 1
Sexual orientation SEXOR 0: heterosexual
1: homosexual
2: bisexual
Flexible, exogenous variables
Living Agent ALIVE 0: dead
1: alive
Mother MO ID of mother agent
Father FA ID of father agent
Age of agent AGE Age of agent in years
Mortality DEATHP Age-dependent death probability
Flexible, endogenous variables
Pregnancy PREG 0: not pregnant
>0: Day of pregnancy
<0: Days of infertility (after birth or life-long)
Number of children NOB Number of births in the life-time of a female agent
Partner ID PID 0: no partnership
>0: ID of partner
Sex of partner PSEX 0: female
1: male
Begin of current relationship BCR <0: No. of iterations until next partnership
0: no partnership so far
>0: iteration of begin of current partnership
Number of partners NOP Number of partnerships in the life-time of the agent
Condom usage CONUSAGE Probability of agent to use condoms
Pill usage PILLUSAGE Probability of female agent to use oral contraceptive
One night stands NSM Number of sexual contacts outside partnerships (single and unfaithful behavior)
Syphilis SYPH 0: not infected
1-9: different states of syphilis
Duration of syphilis SYPHDUR Duration in the current state of syphilis
Infector SYPHINFECTOR ID of agent who infected the agent with syphilis

Table: Overview of agent memory variables used in the SILAS-model.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: