3 ms·
Sorry for the pendantry: it sounds like you mean linear interpolation, rather than extrapolation. Extrapolation is where you project outside the range of previo
by quietbritishjim 6y ago
Sorry for the pendantry: it sounds like you mean linear interpolation, rather than extrapolation. Extrapolation is where you project outside the range of previously collected data, and is particularly likely to result in false conclusions. Interpolation is where you estimate a value inside the range of collected data from points around it, and is usually a lot less dangerous. But I agree with your point that, in this case, it does seem pretty baseless.
- crazygringo 6y agoBut the article is extrapolating, not interpolating. Extrapolation is when you're extending to outside of the measured data range. In this case, the measurements were all of large values, and it's extrapolating (linearly) to small values. Just because the values being extrapolated to are smaller, rather than larger, doesn't make it interpolation. Which is precisely why this has the risk of false conclusions, like you say.