Member Avatar for spe_eddy

Sorry - I know this isn't strictly Python, and i'm more asking for a formula than code!
Usually Excel functions show how they are calculated but i can't find anything on this one!
So given a set of data, predict what the next value will be based on the previous values
Thanks

  • 3 Contributors
  • 6 Replies
  • 465 Views
  • 21 Hours Discussion Span
  • Latest Post Latest Post by spe_eddy
Member Avatar for nezachem

Excel statistical functions: TREND

The TREND(known_y's, known_x's, new_x's, constant) function is used to perform Linear Regression. A least squares criterion is used and TREND tries to find the best fit under that criterion.

Is that what you are looking for?

Member Avatar for spe_eddy

not sure - Is that the definition from Excel?
So trend just uses the least squares criterion?

Member Avatar for griswolf

No. It uses linear regression which (waving my hands broadly) amounts to finding the 'best possible' line through a set of points, and, given N-1 coordinates for a new point, guessing that the other coordinate puts the point on that line. With error bars. Least squares is a good but not unique way to decide where that 'best possible' line lies.

¿Shouldn't this thread be in the Computer Science forum?

Edited by griswolf because: n/a
Member Avatar for spe_eddy

Ok thanks, would you rec recommend any other methods or do you think i should use least squares?
And yes it probably should be in Computer Science, sorry

Member Avatar for griswolf

Least squares has several things going for it:

  • It is very common, so there are many examples, code snippets, etc
  • It is well behaved
  • It is reasonably simple

Unless I had a reason to suppose something else would be better, I'd default to least squares fitting.

Warning: All these things work, if at all, only on 'good' data. Statistics and various curve fitting techniques are prime candidates for GIGO problems.

Edited by griswolf because: n/a
Member Avatar for spe_eddy

Thanks very much,
do you know how i'd find the best fit line in Python though? Is it easily writeable?
My variable are independent of each other though - i am trying to predict the mean temperature of a day given the previous days' mean temperature, so the day is irrelevant - it's only the previous temperatures i'm concerned with.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.