Why is the answer "No" for the first question
Hi,
Can you elaborate why is the answer No for the first question? cannot we conduct a regression analysis if there is a relationship between cones orders and sales of ice-cream?
imo there is no any form of propably causality. Beside trivial case (order 0 cones, sell 0 icecreams etc)
Hi - I think causality is the key here. There is a relationship between the two events, however the one event occurring does not necessarily cause the other event to occur. For instance, ordering a certain number of cones will not cause a certain number of ice creams to be sold. In other words, the ordering of a certain number of cones does not mean that a certain number of ice creams will be sold as other factors (daytime temperature, shop location, flavors of ice cream) cause ice cream to be sold.
There is a dual causality relationship between ice cream sold and cones purchased.
Let's use extreme examples to make it easier to understand: If I order 1 cone, I cannot sell more than 1 ice cream, even if it's during a HOT day. There is a cause there.
If I order 100 cones, selling ice cream will depends on multiple factors, such as temperature, but also price. Let's say I made a clerical error and ordered 1000 cones. I will most likely lower the price, or offer a 2-for-1 rebate. This would CAUSE an increase in ice cream sold.
There is also a causality effect the other way around. On most Just-in-Time inventory management, number of ice cream sold will determine how much cones I should purchase on my next batch.