leonardoldc
New member
- Joined
- Apr 4, 2022
- Messages
- 1
I have the below dataset and i need to classify to a dealership the most under valued cars (most cheap cars but beyond price considering variable KM also).
I thought to calculate the price ratio (price/price average) and km ratio (km/km average ) and then classify the must under valued cars by lowest (km ratio * price ratio) value, but it do not work well because for example a price ratio of 0,9 is better than a km ratio of 0,9, because price varies less than km. You can see the below example, both cars have final ratio of 0,9 but in real world car 1 is most under valued than car 2 because price ratio its more important.
car 1:
price ratio: 0,9
km ratio: 1,0
final ratio: 0,9 * 1,0 = 0,9
car2:
price ratio: 1,0
km ratio: 0,9
final ratio: 1,0 * 0,9 = 0,9
So it seens to me price should have more weight in the calculation, but i dont know how to deffine the fair weight to price and to km, maybe something using their standard deviation.
So my question is, how would you calculate to classify by the cheapest cars?
*The price and km averages and their SD it's calculated by me based on other dataset, so if need to get data from the sample population to create other metrics or calculations, its possible.
I thought to calculate the price ratio (price/price average) and km ratio (km/km average ) and then classify the must under valued cars by lowest (km ratio * price ratio) value, but it do not work well because for example a price ratio of 0,9 is better than a km ratio of 0,9, because price varies less than km. You can see the below example, both cars have final ratio of 0,9 but in real world car 1 is most under valued than car 2 because price ratio its more important.
car 1:
price ratio: 0,9
km ratio: 1,0
final ratio: 0,9 * 1,0 = 0,9
car2:
price ratio: 1,0
km ratio: 0,9
final ratio: 1,0 * 0,9 = 0,9
So it seens to me price should have more weight in the calculation, but i dont know how to deffine the fair weight to price and to km, maybe something using their standard deviation.
So my question is, how would you calculate to classify by the cheapest cars?
*The price and km averages and their SD it's calculated by me based on other dataset, so if need to get data from the sample population to create other metrics or calculations, its possible.