In many situations it is not possible to know all of the variables involved.
Let's answer this question by looking at a few different systems and imagining which variables might be the most impactful.
a sports player
If I wanted to create a model of a sports player what variables might I choose? I want to choose a variables that are meaningful. Maybe:
- number of injuries
- number of points scored
- number of games won
- number of games lost
Limitation: Of course if I haven't considered their current health, current ability, and their strength compared to a certain team, I would not have an accurate model.
I think these variables MIGHT help me model a sports figure.
Maybe I wanted to create a model of a country. The variables I choose might be related to what I want to do with my model.
- percentage of population with electricity access
- electricity - consumption
- electricity - exports
- electricity - imports
These variable are used from the CIA world factbook with gratitude. I wonder why they create models of EVERY COUNTRY ON EARTH?
Limitation: I wonder what happens to electricity when the gas is suddenly cut off? Or if the government introduces a huge subsidy for renewable energy? I would not have an accurate model.
I wonder how these variables might inform my choices about buying a car:
- total cost
- monthly average gasoline cost
- monthly repair cost
- monthly resale value
Limitation: I didn't include number of unscheduled repair visits nor did I include number of repair recalls or warranty visits.
- Describe the limitations of computer (mathematical) models.