Calculating of LOD(DL) based on the Calibration Curve

[bloodstone001]

Q2B gives us a formulation to the calculation of DL : DL = 3.3 σ/S.
Thereof, s(σ)could be estimated based on the calibration curve, that is, as Q2B said " a specific calibration curve should be studied using samples containing an analyte in the range of DL. The residual standard deviation of a regression line or the standard deviation of y-intercepts of regression lines may be used as the standard deviation."
But, I don't know how to calculate.Would somebody kindly give me a example?

[mbulentcakar]

hi,
I'll try to be pragmatic. I hope you can make a regression data analysis with MS Excel (you should have select "analysis toolpak" option in Tools>Add-Ins before). Any other stat software would be able to make a regression analysis.
When you select "data analysis" from "tools" excel will ask you to select an analysis from a list. You can select regression and give the y and x range. Then you can see the residual standart error.
Good luck.

[ntruong]

s(σ)could be estimated based on the calibration curve

It is true that you could estimate the limit of detection, but I think you still have to prove it by doing the actual experiment to see whether or not your method can deliver.

[bloodstone001]

Thanks. It is of great help for me!

[michaldousa]

If you have an interest in calculation formula in Excel I would send you (need E-mail). It is not simple like it may seem.

[yangz00g]

oops!