2
$\begingroup$

I have read related questions about linear regression models with hour of day variable, but they didnt help me at all, so I want to describe my doubt exactly:

I am facing a linear regression problem, where I have a several variables just to predict another related with pollution degree. I have doubts with two of these variables, they are: "hour of day" and "number of cars in these hour". Before the model, just thinking about the physical sense of the problem, we can think that simply the hour of the day cannot be related with the pollution at all, the hour variable is more related with the number of cars, the number of cars increases in day hours and decreases at night. Am I doing the regression model right if I include the numbers of cars variable, knowing that this variable is related with the hour of day in which the pollution variable was measured?

$\endgroup$

1 Answer 1

1
$\begingroup$

Hour of the day might be directly related to pollution if heat is related to pollution, which probably depends on exactly what sort of pollution you are measuring.

But hour is certainly a sort of marker for other things that are related to pollution - yes, number of cars, but also all sorts of other activity that could change pollution levels such as amount of fuel being burned.

The exact relation is going to depend on the particulars of your case.

$\endgroup$
2
  • $\begingroup$ Yeah, your right in everything you have said, but if I only use these two variables to measure pollution? I mean, when I make the regression model, I only include "numbers of cars " variable, but does it make sense knowing that this variable has been measured in a specific hour of the day? For example: Measure 1: 100 cars ( hour: 11:00 am) Measure 2: 400 cars (Hour 2:00 pm) Measure 3: 20 cars (Hour 3:00 am ).... $\endgroup$
    – DannW
    Commented Dec 25, 2019 at 13:05
  • $\begingroup$ All the combinations of cars and hour make some sort of sense. They each answer different questions. $\endgroup$
    – Peter Flom
    Commented Dec 25, 2019 at 13:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.