Skip to main content
Question

Least Squares Adjustment


deanhowell
Influencer
Forum|alt.badge.img+23

Has anyone ever used FME to undertake a Least Squares Adjustment on survey data?

2 replies

mark2atsafe
Safer
Forum|alt.badge.img+44

I think the Affiner or AffineWarper are the transformers that you can use here.

if you have vectors by which to adjust the data, then the AffineWarper transformer uses least squares to calculate the affine coefficients. The Affiner I guess you would use if you already had those coefficients.

One limitation that I'm aware of is lack of output concerning residuals. I don't think they appear in the log and there is no way to set a maximum residual. So you adjust the data using least squares, but you don't have any measure of how good the adjustment was.

Another tool you might try is the RCaller transformer. I'm 100% sure it will do least squares calculations, though whether it will do survey adjustments I'm not certain. It might be worth a look at the documentation for R to see if it can help.

I hope this is of use to you, and hopefully some users on this forum will have some more practical advice. If you do need more information about the residuals and errors, and are in touch with our support team, then suggest to them that FMEENGINE-15298 is a good place for them to start investigating.


deanhowell
Influencer
Forum|alt.badge.img+23
  • Author
  • Influencer
  • July 31, 2019
mark2atsafe wrote:

I think the Affiner or AffineWarper are the transformers that you can use here.

if you have vectors by which to adjust the data, then the AffineWarper transformer uses least squares to calculate the affine coefficients. The Affiner I guess you would use if you already had those coefficients.

One limitation that I'm aware of is lack of output concerning residuals. I don't think they appear in the log and there is no way to set a maximum residual. So you adjust the data using least squares, but you don't have any measure of how good the adjustment was.

Another tool you might try is the RCaller transformer. I'm 100% sure it will do least squares calculations, though whether it will do survey adjustments I'm not certain. It might be worth a look at the documentation for R to see if it can help.

I hope this is of use to you, and hopefully some users on this forum will have some more practical advice. If you do need more information about the residuals and errors, and are in touch with our support team, then suggest to them that FMEENGINE-15298 is a good place for them to start investigating.

Thanks @mark2atsafe I will investigate these transformers further.


Reply


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings