I'm currently struggling with a question involving probability and statistics.

I have this dataset of sales, and I was trying to make a probability of sales based on that dataset and the data that it provides me of weeks, months and years back. I started using Bayes Theorem to do it, and after a conversation with a data scientist, it was suggested that I applied concepts like prior and posterior distribution to my model. The thing is, those concepts are way ahead of my statistical knowledge right now, so I would have to take time and study them so that I could move on with my project.

But, I thought to my self, since I've got a whole dataset containing all the information I'm gonna need (the population information) there's no need to try and use such statistical concepts, I could just use the dataset observations to create the probability, since everything I could try to predict is already there, right? In my head, those concepts would only need to be used if I was working with samples.

Can someone clarify this to me?

  • 1
    $\begingroup$ You say you have the population, but I'm guessing you want to use this information to make a prediction about the future. Do you have the population of future values, or do you want to make inferences about the future based on what you've observed? $\endgroup$
    – Dave
    Jul 15, 2020 at 16:59
  • 1
    $\begingroup$ That's right, I wanted to use informations from the past to predict informations about the future. $\endgroup$
    – Caldass_
    Jul 15, 2020 at 17:04


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.