Hello, I'm new to these forums and I have a question regarding predicting data. I've very new to statistics as I'm a freshmen in college and I am trying to develop a model that best predicts future data based on a linear set of past data (say 3-6 numbers) I don't want the model to simple predict based on what "fits" past data, but is more or less accurate for predicting future data. I have my data set which consists of 100 sets of 3-6 numbers, which are organized by year (say I have 4 numbers, I'd have the "year 1" number, "year 2" number, etc) and so I'd predicting based on that. it's all in excel (which I'm also new to) and I feel kind of lost right now.

I'm not asking for someone to just give me a formula, but I really want to understand what the reasoning is behind statistical forcast. If anyone could help me, I'd really appreciate it. I apologize if my question is ambiguous, I'm still not familiar with the "lingo"

Thanks in advance!