Accuracy vs Precision vs Resolution issues

[maxschommer]

Here it is said that the narrowly distributed temperature readings represent a more accurate thermometer, when I think it is intended to mean a more precise one. The book previously established that we will assume no systemic bias, or inaccuracy. This occurs in other places before this, and perhaps in future sections of the book. If I track down more instances of this I'll add them as comments to this issue.

Kalman-and-Bayesian-Filters-in-Python/03-Gaussians.ipynb

Line 1188 in 9e3d2f6

"If we think back to the thermometer, we can consider these three curves as representing the readings from three different thermometers. The curve for $\\sigma^2=0.2^2$ represents a very accurate thermometer, and curve for $\\sigma^2=1^2$ represents a fairly inaccurate one. Note the very powerful property the Gaussian distribution affords us — we can entirely represent both the reading and the error of a thermometer with only two numbers — the mean and the variance.\n",

[maxschommer]

Kalman-and-Bayesian-Filters-in-Python/03-Gaussians.ipynb

Line 1003 in 9e3d2f6

"What does this curve *mean*? Assume we have a thermometer which reads 22°C. No thermometer is perfectly accurate, and so we expect that each reading will be slightly off the actual value. However, a theorem called [*Central Limit Theorem*](https://en.wikipedia.org/wiki/Central_limit_theorem) states that if we make many measurements that the measurements will be normally distributed. When we look at this chart we can see it is proportional to the probability of the thermometer reading a particular value given the actual temperature of 22°C. \n",

Here is another instance. I think in context that "precise" is meant instead of "accurate". While it is a true statement that no thermometer is perfectly accurate, the variance from reading to reading that is being alluded to here is a precision issue, not necessarily an accuracy one.

[maxschommer]

Here "precision" is used when I think "resolution" should be used. "precision" is meant to convey how tightly together readings are clustered, where "resolution" is the minimum delta that a sensor can display.

Kalman-and-Bayesian-Filters-in-Python/03-Gaussians.ipynb

Line 1019 in 9e3d2f6

"In practice our sensors do not have infinite precision, so a reading of 22°C implies a range, such as 22 $\\pm$ 0.1°C, and we can compute the probability of that range by integrating from 21.9 to 22.1.\n",

[maxschommer]

https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python/blob/9e3d2f6ed023d937587cf2ef2ecfbf7afc3d8054/04-One-Dimensional-Kalman-Filters.ipynb#L639

https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python/blob/9e3d2f6ed023d937587cf2ef2ecfbf7afc3d8054/04-One-Dimensional-Kalman-Filters.ipynb#L645

I think "accurate" should be replaced with "precise" in these two instances as well.

[maxschommer]

https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python/blob/9e3d2f6ed023d937587cf2ef2ecfbf7afc3d8054/05-Multivariate-Gaussians.ipynb#L852

Perhaps instead of "error" it should be "confidence" or "uncertainty"