I'm reading this chapter forecasting principles and practise from a forecasting book.
The author has explained a linear regression model. Now this linear regression model will definitely have some errors in fit. For these errors, the author has listen certain characteristics, which should be taken into account. He assumes that these errors:
- Have mean zero; otherwise the forecasts will be systematically biased.
- Are not autocorrelated; otherwise the forecasts will be inefficient as there is more information to be exploited in the data.
- Are unrelated to the predictor variable; otherwise there would be more information that should be included in the systematic part of the model.
The problem is that I'm not able to comprehend these and come up with an example to get a hang of these points. Could somebody point out some examples / situations from daily life so as to understand these points. For example the author wrote in point 2.] ..."there is more information to be exploited in the data"...
What is this information? How does it come into being? If I don't even know how this information comes into existence, then how will I be able to understand its existence or effects for that matter?
Is there a simple explanation with example for my question? I have been trying to understand this since quite some time, please help.