Help on what to learn about

Users
#1

Hi everyone!
I hope you can help me. I would like to know which subject area to learn about given my project. I have done lots of searching but can’t seem find the exact topic.

I am working with some time series weather data. Rather than “forecasting” the next result each time a new data point is entered I would like to find patterns in data that lead to a high probability result.

For example a successful pattern would be in 90% of instances, every time we have 5 days of plus 30 degree heat the 6th day is 28 degrees and above.
Or in 75% of instances of 3mm rainfall, the following day always had 1mm or above.

Hopefully you can see what I am trying to do. Rather than forecast each day, I want to identify patterns that lead to relatively high probable forecasts. So I wouldnt have a new forecast every day, far from it, once a week or month “the stars would align” and we could make a pretty solid forecast for the following day.

Sorry, long post but hopefully you can point me in the direction of the areas I need to study in order to create something of real worth.

Thanks very much

#2

Hi Sam,

Your question isn’t really to do with Python.

What you describe is forecasting.

One approach is to look at learning about statistics, in particular

“time series”. Another area to look at is “machine learning”. There will

have been a tonne of work already done in this area by banks for

financial markets, although a lot of that will be proprietary and

secret. Another area that does something similar is economic modelling.

Some other keywords you can look for are “probabilistic”, “Baysian”,

and “modelling”.

#3

Thank you Steven, that’s very helpful.