1–10 of 16 posts
Show HN: An easy-to-use online curve fitting tool
byx2000.github.io
It'd be nice if there was some demo data because I might want to play with it to see how it works, but don't have any data to use it on.
The fit diagnostics at the top of the plot are inadequate. This needs at a minimum error estimates on the estimated parameters (probably bootstrap) and ideally some kind of "error envelope" on the plot.
For example, suppose you measure the decay of a radioactive source at fixed times t = 0,1,2,... and fit y = A e^{-kt}. The only randomness is small measurement error with, say, SD = 0.5. The bootstrap sees the huge spread in the y-values that comes from the deterministic decay curve itself, not from noise. It interprets that structural variation as sampling variability and you end up with absurdly wide bootstrap confidence intervals that have nothing to do with the actual uncertainty in the experiment.
The fit diagnostics at the top of the plot are inadequate. This needs at a minimum error estimates on the estimated parameters (probably bootstrap) and ideally some kind of "error envelope" on the plot.
I don’t think you can do anything sensible here without making much stronger modelling assumptions. A vanilla non-parametric bootstrap is only valid under a very specific generative story: IID sampling from a population. Many (most?) curve-fitting problems won't satisfy that. For example, suppose you measure the decay of a radioactive source at fixed times t = 0,1,2,... and fit y = A e^{-kt}. The only randomness is s…
The fit diagnostics at the top of the plot are inadequate. This needs at a minimum error estimates on the estimated parameters (probably bootstrap) and ideally some kind of "error envelope" on the plot.
I don’t think you can do anything sensible here without making much stronger modelling assumptions. A vanilla non-parametric bootstrap is only valid under a very specific generative story: IID sampling from a population. Many (most?) curve-fitting problems won't satisfy that. For example, suppose you measure the decay of a radioactive source at fixed times t = 0,1,2,... and fit y = A e^{-kt}. The only randomness is s…
Maybe a residuals plot and IID tests of residuals (i.e. tests of some of the strong assumptions!) would be a better next step for the author than error estimates, but I stand by my original feedback. Right now even the simplest case of a straight line fit is reported with only exact slope & intercept (well, not exact, but to an almost surely meaningless 16 decimals!), though I guess he thought to truncate the goodness of fit measures at ~4 digits.