The Quattro Sensor and Luma

Started Jul 18, 2014 | Discussions thread
Joofa Senior Member • Posts: 2,791
Re: Response Spreadsheet and Curves Posted
1

Eric Fossum wrote:

JimKasson wrote:

Aside: I minimized the sum of the squared error, but minimizing the square root of that or the rms error would all yield the same result.

Are you sure about this? When you square the error the impact of deviations is increased and hence strong deviations play a larger role in determining the minimum-error solution.

Besides, square root gets into issues with negative values.

What Jim means is that sum_i(error_i^2), i.e., SSE, sqrt(sum_i(error_i^2), and sqrt(1/N*sum_i(error_i^2)), i.e., RMS, would yield the same solution, because sqrt is a monotonically increasing function.

I guess, you took that to mean sum_i(sqrt(error_i)), which is different.

-- hide signature --
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow