Use Excel to find the similarity between functions.

I am writing a paper and I came up with a function that uses a hypothetical relationship to predict the value of one variable at different points in time, I graphed it, and then graphed the actual readings from an experiment. How can quantitatively describe how close the two trends are? In other words, how can I quantitatively describe the amount of support the hypothetical relationship get from the experiment? See image for more ideas, predictive function gives values in blue, actual readings are orange.
Screenshot 2015-09-15 21.39.28.png
 

DrClaude

Mentor
6,969
3,141
You could perform a χ2 test. You can also calculate the R2 for your model, since you already have the value for the linear regression.

If you don't know what any of that means, you can start with https://en.wikipedia.org/wiki/Goodness_of_fit
 

FactChecker

Science Advisor
Gold Member
2018 Award
5,086
1,791
You can get the confidence intervals for the slope and intercept of the regression line. For a given confidence level, your theoretical parameters will be in or out of the confidence interval. Just eyeballing the data, the regression line, and your theoretical line, I think there is something missing from your theory. I think it is likely to be statistically rejected.
 

Related Threads for: Use Excel to find the similarity between functions.

  • Posted
Replies
9
Views
4K
Replies
3
Views
90K
  • Posted
Replies
3
Views
3K
Replies
1
Views
5K
  • Posted
Replies
3
Views
63K
  • Posted
Replies
3
Views
2K
Replies
1
Views
538
  • Posted
Replies
3
Views
23K

Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving
Top