How awful is that! You just find the correlation coefficient of x and you get the lagged value of the function associated with it! That doesn’t look so good does it? And in fact there are many instances where you have actually measured the correlation, but only found it to be negative or insignificant. So how do you find out whether the correlation is real?
Well, for one thing, it is not as easy as measuring the correlation coefficient between ice cream and milk. Suppose you want to calculate the correlation between the price level and the productivity level of a service station, how are you going to do it? First you have to go down to the local ice cream store and get some milk and ice cream. Then, measure the productivity level in the future, and when you get back to the store you find out that the productivity has decreased.
You might say “What now?” Well, I say “Okay, first find out the values of all the ice cream you bought, then find out the productivity levels of the ice cream shops in your local.” This is a rather simplistic example, but I am sure you get the picture. You take all the relevant data, and you find out that they are all close to each other. The correlation is very high, and therefore there is a strong relationship. So, you can conclude that the correlation coefficient is real.
But what about a negative correlation? Can you have a negative correlation, where there is no positive correlation at all? Well, think about the last time you went to the ice cream store, and you saw that the line was empty. Did you say “panic button” and take your leave? That kind of thing is called a zero correlation.
Now, what can you do to improve the correlation? One thing you could do is increase the number of people who come into the store and decrease the number of people who leave. Obviously, that will not change the fact that you are increasing the amount of labor that needs to be done in order to make the money. However, it can lower the overall efficiency of the operation. If the customers do not spend as much time in line waiting to buy ice cream, then you will be spending more time waiting in line to buy ice cream, and the production efficiency will fall.
Now, let us look at a negative correlation. What can you do to improve the negative correlation? Well, you need to get more people into line to buy your products, but you also need to get them to come up through other channels. This can be accomplished by offering coupons and special deals. These can help people get through the process faster, while at the same time improving the positive correlation between the number of customers and the amount of ice cream sales.
So, in conclusion, the correlation between ice cream sales and the amount of labor that are spent in an operation does exist. It is up to you to test this correlation and see if there is indeed a negative correlation. If there is a negative correlation, then you should consider changing the way that you are operating your business in order to improve the positive correlation. If you find that there is a positive correlation, then keep doing what you are doing, and just tweak it a little bit. You will be pleasantly surprised by the results.