0

I have recently come across some papers in my field. They had a large (X,Y) data set, it was binned into 4 bins. Least square method was used to find slope (m) and y intercept (b). In the table that contains the slope and y intercept they include what they call "1 standard deviation error" for each of the m and b. My question is what is this error and was it obtained?

Link to paper Second page table 1 and figure 1

https://arxiv.org/abs/2111.05544

Axxxxx
  • 31
  • 1
    Can you link the paper? That would help – Apprentice May 28 '22 at 15:15
  • It sounds like they are referring to the standard errors of the estimates for the intercept and slopes. These are standard outputs from statistical software and allow you to compute confidence intervals for these parameters, as well as obtain p-values for them. See https://stats.stackexchange.com/questions/85943/how-to-derive-the-standard-error-of-linear-regression-coefficient – fmtcs May 28 '22 at 15:27
  • I updated the question – Axxxxx May 28 '22 at 15:29
  • 1
    Although the paper doesn't appear to explain its fitting method, the figure makes it clear these are ordinary least-squares fits (for the logged variables). The "$\sigma$ values" would of course be the standard errors as routinely output by any least-squares fitting algorithm. – whuber May 28 '22 at 15:33
  • How are these standard errors computed for the slope and intercept? – Axxxxx May 28 '22 at 15:36
  • 1
    @Axxxxx see the link in my previous comment – fmtcs May 28 '22 at 15:38
  • 1
    https://stats.stackexchange.com/search?q=compute+regression+standard+error. – whuber May 29 '22 at 12:20

0 Answers0