# Regresion analysis [statistics]

Hello, I have a problem and I need advice or help. I collected data about the parking lot, such as its occupancy, what is in it and how it works in general in principle. Well, when my teacher told me later that I didn't have enough data and that I should have done the analysis more often to get how many spaces were occupied in the parking lot, only option left was regresion analysis. How could I theoretically, with the help of some analytical tools and methods, or with the help of mathematics, calculate the occupancy of those places? As I wrote, I have data for a short period, about three weeks, maybe more, and I have information on how many vehicles there were for the entire semester, and also how many vehicles there were during my measurement/analysis. If random effects are also to be counted, I can add that something was being renovated in the parking lot all the time, some days have more vehicles than others, and I measured from 8:00 a.m. to 11:00 a.m. and 3:00 p.m., each measurement lasted one hour . If you have any questions, write to me. Excel is in slovak language, its messy but if you will look at it i have everything zoned, how many free space i have and how many cars are there, and also how much cars went throught ramps on which position. I tried to do regresion but i dont know if i was sucesful.

I am sorry for my english, its not my frist language. If you write me massage i can give you money.

## 1 Answer

You should a priori try an additive or multiplicative model, for a time series with some trend and some periodic component. See for example https://www.abs.gov.au/websitedbs/d3310114.nsf/home/time+series+analysis:+the+basics.

You get other results, in particular YouTube videos, if you google "time series trend multiplicative model". Hope that helps.

PS: I didn't mean to submit this as answer but rather as comment (actually, to correct "tend" => "trend" w.r.t my previous comment), but it was saved as answer so I thought I leave it...

- 1 Answer
- 73 views
- Pro Bono

You should a priori try an additive or multiplicative model, for a time series with some tend and some periodic component.