The Student Room Group

Interpreting regression results.

I've just done a linear regression analysis in Microsoft Excel and now I need to interpret my results to determine whether the linear model is suitable for this data. But I'm not entirely sure what I'm meant to be looking for. Can anyone point me in the right direction? :confused:
(The results include things like R Square, t-stats, p value and the residuals)
Reply 1
What level of stats are you at?
Reply 2
I'm doing the 1st year of my degree, but the stats is only part of a module. I'm doing accounting.
Reply 3
Look up the different results in wikipedia, and interpret from there. The definitions will form an obvious basis for your answer
Reply 4
From my R-square results it suggests it's a good fit but the p-value is really low.
So I tried a double log relationship and the R-square improved and it looks like a better fit graphically but the p-value is still really low (less than 0.05). Does this matter? All the examples I have also have low p-values...
Reply 5
Probably talking to myself here but I've thought through it and realised that I think I have H0 and H1 confused. In which case the low p-value would be right since I would reject H0 and conclude that there is evidence that there is a linear relationship between x and y.
Just wanted confirmation that I'm right :tongue:

Quick Reply