Databases
StackOverflow2013
Postsdbo
POLinear fit with a previously known slope
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POLinear fit with a previously known slope
    primarykey
    data
    text
    <p>I'm trying to fit a linear function for whom I know the slope to this data in R:</p> <pre><code>&gt; flN eCenter eLow eHigh arrivalTime_4_8 timeError_4_8 vCenter vLow vHigh 1 56.4997 60.25967 52.73972 2012-05-17 02:01:44 41.01283 2.298964 2.236939 2.367869 2 105.5787 108.02476 103.13254 2012-05-17 01:57:37 27.06803 1.787009 1.771748 1.802860 3 164.0613 166.50385 161.61884 2012-05-17 01:49:47 60.42015 1.530334 1.522997 1.537860 4 226.9886 229.78374 224.19346 2012-05-17 01:49:20 50.83871 1.386016 1.381233 1.390904 </code></pre> <p>My Y is <code>arrivalTime_4_8</code> and the error in Y is <code>timeError_4_8</code>. My X is <code>vCenter</code> and the error <code>vHigh</code> and <code>vLow</code> for the upper and lower values.</p> <p>So, have 3 problems about this (I'll put them in order of importance):</p> <p>1 - How do I fit a linear function with a set slope? I tried to use <code>lm</code>, but I can't find a way to put the slope there:</p> <pre><code>lm(formula = arrivalTime_4_8 ~ vCenter, data = flN) </code></pre> <p>2 - I know I must use weights = 1/sqrt(error) to get a weighted fit, but how do I do that with a timeframe axis?</p> <p>3 - How do I calculate the weights to use if I have the error measurements in 2 axis? A simple squared sum?</p> <pre><code>&gt; dput(flN) structure(list(eCenter = c(56.4996953402131, 105.578651367445, 164.061343347697, 226.988601498086), eLow = c(60.2596687117468, 108.024759195324, 166.503850666149, 229.78374170578), eHigh = c(52.7397219686794, 103.132543539565, 161.618836029245, 224.193461290392), arrivalTime_4_8 = structure(c(1337216504.88343, 1337216257.43636, 1337215787.45439, 1337215760.18075), tzone = "", class = c("POSIXct", "POSIXt")), timeError_4_8 = c(41.0128309582924, 27.0680250428967, 60.4201539087451, 50.8387114506802), vCenter = c(2.29896400265163, 1.78700866407098, 1.53033400915538, 1.38601563560752), vLow = c(2.23693860876912, 1.77174836673712, 1.52299693760316, 1.38123278889559), vHigh = c(2.36786898717605, 1.80286021104626, 1.5378599964982, 1.39090403398638)), .Names = c("eCenter", "eLow", "eHigh", "arrivalTime_4_8", "timeError_4_8", "vCenter", "vLow", "vHigh"), row.names = c(NA, -4L), class = "data.frame") </code></pre>
    singulars
    1. PO
      singulars
      1. PTAnswer
    1. This table or related slice is empty.
    1. PTQuestion
    1. USjbssm
    1. USjbssm
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POLinear fit with a previously known slope
      1. This table or related slice is empty.
      1. VTUpMod
    1. COPlease clarify: (1) are you just trying to estimate the intercept (and its uncertainty), given that you know the slope? (2) can you clarify this question? I don't understand ... (3) errors-in-measurements models are a little bit complicated and probably can't be easily fitted with `lm`.
      singulars
      1. POLinear fit with a previously known slope
      1. USBen Bolker
    2. CO1 - Exactly, just the uncertainty of the fit and the intercept. 2 - More simply put, how do you estimate errors for an axis with times (do I have to convert everything to seconds for instance or does R do that automatically). 3 - I know how to use the error for fitting in just one axis (for instance if only the Y axis had error I would just give an weight vector to lm equal to 1/sqrt(error)), I'm not sure how to do it for errors in both axis tough.
      singulars
      1. POLinear fit with a previously known slope
      1. USjbssm
    3. COcan you please post the results of `dput(flN)` in order to make it a little easier to answer the question (i.e. make it reproducible)? I believe R will convert the dates to seconds automatically. Again, dealing with errors in X is non-trivial.
      singulars
      1. POLinear fit with a previously known slope
      1. USBen Bolker
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload