What do you want to explain? What is your research hypothesis, or do you just want to do data mining?

Carlos Jimenez-Gallardo I have a magnetoresistance curve for which I have to perform statistical analysis. I want to know how to do that ?

?????

• To fit the R(H) curve to a physical or empirical model
• To study how resistance varies under repeated measurements at the same H
• To compare the magnetoresistance of two or more materials under the same conditions
• To compare the field sensitivity between different materials

Carlos Jimenez-Gallardo Ok.

• Can you please shed some light on "To fit the R(H) curve to a physical or empirical model" ?

Based on my understanding of the problem, there can be three general cases: classical magnetoresistive response, ordinary magnetoresistance, and the one related to ferromagnetic materials. Therefore, there might be a linear, quadratic, or non-linear model (which involves saturation).

You can check:

• Linear models using R or Excel: https://rpubs.com/mbounthavong/linear_regression_using_R
• Non-linear models: https://rpubs.com/abbyhudak/nonlinreg
• Models with saturation (related to ferromagnetic behavior) https://rpubs.com/JackJWright/deviance

I hope this helps

greetings

If a magnetoresistance curve is U-shaped, use a quadratic fit (2nd order polynomial). If it has a bell-shaped peak, a Lorentzian or Gaussian fit works best. If the curve is wavy and more complex, consider using a higher-order polynomial fit.

Veronika Burobina Many thanks for your reply. Can you please share the details on how to the fitting of the MR curve and open-source softwares ?

Origin is one of the best tools for curve fitting. In Origin, in Column A, enter magnetic field values (H); in Column B, enter resistance values (R). Highlight both columns. Plot the data: Basic 2D → Scatter. What is the shape of your curve? One should choose a fitting method based on the shape of the curve.

Click the plot to select it, then go to Analysis → Fitting and choose a fitting method based on the shape (Polynomial Fit or Nonlinear Curve Fit).

Origin is not open source, but it is widely used in research. If you are comfortable with coding and prefer an open-source alternative, Python is a great choice.

Veronika Burobina Many thaanks for the knowledge on Origin. I know it already. Can you please tell me about the equations to be used for fitting ?

What does your curve look like? What material are you working with? A polynomial fit, such as R(H)=aH^2+bH+c, is commonly used MR curves, but not always. If your curve has a sharp peak, a Lorentzian or Gaussian model may be more suitable. If there is hysteresis, you should analyze the increasing and decreasing branches separately...