Use variable from previous step (lag variable)

[AlexandraGlaz]

Hi! I'm building macroeconomic model and some of my variables have lagged variables, for exampe AR(1) prosses. So to do it now I create additional var for each lag in Class, save var on step t-1 and using in on step t. So that I have a lot of additional vars.
Is someone know how to do in more effective?
I will be grateful for any help!

Please find my current code in attach.
mesa_code.txt

[rht]

Can you edit your question with a minimum self-contained code that can be built on, for a concrete starting point?

[AlexandraGlaz]

Done

[rht]

Try this code:

import mesa


class Household(mesa.Agent):
    def __init__(self, unique_id, model, var1=0, var1_lag=1, var2=0):
        super().__init__(unique_id, model)

        self.var1_timeseries = [var1_lag, var1]
        self.var2 = var2

    def fun1(self):
        self.var1_timeseries.append(self.var1_timeseries[-1] + 1)
        self.var2 = self.var1_timeseries[-1] - self.var1_timeseries[-2]

    def step(self):
        self.fun1()


class Economy(mesa.Model):
    def __init__(self, N):
        super().__init__()
        self.num_agents = N
        self.schedule = mesa.time.RandomActivation(self)
        for i in range(self.num_agents):
            a = Household("hh_{0}".format(i), self)
            self.schedule.add(a)

        self.datacollector = mesa.DataCollector(
            agent_reporters={
                "Var1": lambda agent: agent.var1_timeseries[-1],
                "Var2": "var2",
            }
        )

    def step(self):
        self.schedule.step()
        self.datacollector.collect(self)


model_base = Economy(1)
for i in range(20):
    model_base.step()

agent_data = model_base.datacollector.get_agent_vars_dataframe()
print(agent_data)

[AlexandraGlaz]

It looks easier, thank you