Undergrad Uncertainties from linear interpolation

Click For Summary
The discussion focuses on the challenges of estimating uncertainties when using linear interpolation to model a function of two variables based on simulated data. The original poster seeks to predict values of y from given pairs of (x1, x2) without a known theoretical function, expressing confusion about how to account for model uncertainty in their predictions. Participants clarify that linear interpolation does not provide uncertainty estimates for new data points, as it assumes perfect fits between known points. Alternatives such as polynomial fitting or neural networks are suggested, with the latter allowing for uncertainty estimation through ensemble methods. Ultimately, the conversation emphasizes the need for a clear understanding of the modeling approach to effectively address uncertainty in predictions.
  • #31
I would suggest that you make a simple model and use your techniques to solve it thereby learning your techniques.
 

Similar threads

Graduate Piecewise linear interpolation with uncertainties
  • · Replies 5 ·
Replies
5
Views
2K
Graduate Linear combination of data with uncertainty
  • · Replies 5 ·
Replies
5
Views
2K
Undergrad On pdf of a sum of two r.v.s and differentiating under the integral
  • · Replies 2 ·
Replies
2
Views
2K
Undergrad Collinearity between predictors: what happens under the hood
  • · Replies 13 ·
Replies
13
Views
2K
Undergrad Expected coefficient change from simple to multiple linear regression
  • · Replies 4 ·
Replies
4
Views
2K
Undergrad Sampling theory and random sample
  • · Replies 5 ·
Replies
5
Views
2K
Undergrad Linear Models vs Nonlinear Models
  • · Replies 22 ·
Replies
22
Views
4K
Undergrad Variable Normalization for different variable ranges....
  • · Replies 3 ·
Replies
3
Views
2K
Undergrad Systematic uncertainties in linear fit
  • · Replies 28 ·
Replies
28
Views
3K
Undergrad Randomly Stopped Sums vs the sum of I.I.D. Random Variables
  • · Replies 2 ·
Replies
2
Views
2K