I Does inference help forecasting?

  • I
  • Thread starter Thread starter fog37
  • Start date Start date
AI Thread Summary
The discussion highlights the contrasting priorities between machine learning practitioners, who focus on forecasting, and social scientists, who prioritize inference. Inference involves understanding population parameters and causal relationships through random sampling, while forecasting is concerned with predicting future outcomes. The conversation raises the possibility that a model excelling in inference could also perform well in forecasting, suggesting they are not mutually exclusive. However, it also acknowledges scenarios where a model may be effective at prediction but poor at inference, and vice versa. Ultimately, the relationship between inference and forecasting is complex, with continuity and past behavior often playing significant roles in predictive accuracy.
fog37
Messages
1,566
Reaction score
108
TL;DR Summary
Forecasting vs Inference
Hello,
Many individuals in machine learning/data science are primarily concerned with prediction only (and not inference) while many in the social sciences are mainly concerned with inference only (and don't care about forecasting).

In the case of inference, we consider a population which we want to study and learn about. We collect a random sample and try to understand the underlying parameters that describe the population and search for causal effects between conceptualized variables. Social scientists are all about inference and are not worried about the forecasting performance of their model. On the other hand, many individuals in the machine learning community are focused on forecasting instead and don't worry about checking assumptions, statistical tests of significance, etc. Why not? Is it because the assumptions can be relaxed and we don't run into issues when we deal with lots of data (standard errors are automatically small, etc.)?

I would think that a model that does good inference would also be good at forecasting since forecast is about the future state of the population. If inference is right then forecasting would tend to be right so good inference and good forecasting don't appear mutually exclusive to me.

Or is it possible to have a model with great predicting performance but very poor inference? And a model that does great inference and is terrible at forecasting?

Thank you
 
Physics news on Phys.org
Inference and forecasting can be similar, e.g. if you try to forecast what happens when you change a variable, that's basically a form of inference.

Most machine learning applications have no control over any of the variables, except maybe a single choice of your own. E.g. someone shows up to your website, and you get to show them an ad. The only choice you get to make is which ad you show them. You could pretend to think you know why certain people will click certain ads, but you can't control the environment well enough to verify this explanation.
 
There are situations where forecasting can be much easier and more accurate than inference. I once did a study of which countries to target for increased sales of a product. I considered all sorts of economic, political, and military reasons why a country might want to purchase the product. After trying all sorts of statistical methods, the result was this: By far the best predictor of future sales to a country was the level of past sales to that country and once that was accounted for no other variables had any statistical significance.
It certainly can be true that continuity and habit are the most important factors -- and nothing else matters.
 
  • Like
Likes fog37 and Office_Shredder
I was reading documentation about the soundness and completeness of logic formal systems. Consider the following $$\vdash_S \phi$$ where ##S## is the proof-system making part the formal system and ##\phi## is a wff (well formed formula) of the formal language. Note the blank on left of the turnstile symbol ##\vdash_S##, as far as I can tell it actually represents the empty set. So what does it mean ? I guess it actually means ##\phi## is a theorem of the formal system, i.e. there is a...
Back
Top