Re: [R] Mixed format

2020-01-24 Thread Val
Thank you all for your help. My data has mixed format such as %m/%d/%y,%d/%m/%y,%m-%d-%y,%d-%m-%y etc. and the library (anytime) handles it very well!! Thank you again. On Tue, Jan 21, 2020 at 5:28 AM Rui Barradas wrote: > > Hello, > > Inline. > > Às 09:22 de 21/01/20, Chris Evans escreveu: >

Re: [R] Did the output from summary(lm(...)) change

2020-01-24 Thread Rui Barradas
Hello, No, your memory is wrong, like you say. summary(lm(.)) returns a matrix with the 4th column named "Pr(>|t|)", not "p.value". And this hasn't changed. If you want the p-values, the (old) way to do it still is summary(lm(.))$coefficients[, 4] or, more complicated, summary(lm(.))$coeff

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Paul Bernal
Thanks brother, I really appreciate your help. Have an awesome weekend! El vie., 24 de enero de 2020 5:39 p. m., Richard M. Heiberger < r...@temple.edu> escribió: > I don't have my computer with me, so I am commenting right now on the > visual impression of the email. > > The latitude shows 90,

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Richard M. Heiberger
I don't have my computer with me, so I am commenting right now on the visual impression of the email. The latitude shows 90, 88, ... 2, 89, ... The labels are lexicographically ordered -1, -10, -11,... The latitude binrep look in correct order, and the labels looks like binary in order. These th

[R] Did the output from summary(lm(...)) change

2020-01-24 Thread Dennis Fisher
R 3.6.1 OS X Colleagues My recollection (possibly wrong) is that: summary(lm(YVAR ~ XVAR))$p.value used to return the P value for a linear regression. It does not appear to do so now. Of note: summary(lm(YVAR ~ XVAR)) does report the P value. I realize that I can access the P

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Paul Bernal
Dear friend Richard, Thank you for your interest in helping me through this challenge. As requested, I am providing the two lat and long frames you suggested, plus the one single column I am trying to decode: LatitudeFrame: > dput(LatitudeFrame) structure(list(Latitude = structure(c(90L, 88L, 87

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Richard M. Heiberger
now I am even more puzzled. please complete the following two data.frames and send it to the list. latDegrees lat2Comp -90 -89 ... -1 0 1 ... 89 90 lonDegrees lon2Comp -180 -179 ... -91 -90 -89 xxx

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Paul Bernal
Hi Richard, That was just an example, to show that, for that particular string of binary numbers, the code works as expected. That is absolutely no related to the dataset I provided. If I try the function on the dataset, I get values well over the latitude and longitude boundaries (which should ra

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Richard M. Heiberger
You show the example > fun("10110010") [1] -78 as satisfactory. Where in your posted data set do you find the input string "10110010"? Please post a set of relevant input strings, and the answers you want from them. The rest of the columns are not helpful for this specific exercise. On Fri, Ja

Re: [R] Converting binary number to in Two´s complement representation

2020-01-24 Thread Paul Bernal
Dear friend Rui, Hope you are doing great. Firstly, I want to thank you for your super valuable and kind support of always. As I mentioned in earlier e-mails, I am trying to decode AIS type messages, and the only ones I am having a real hard time with, is with latitude and longitude. I tried the

Re: [R] how to create a plot of permutation of 30 random values and show proportion of values

2020-01-24 Thread David Winsemius
It appears you are not trying to do this within individual SNPs. If I’m wrong then this would need to be done within a grouping procedure. And I’m not at all confident that you can estimate FDRs in this manner, but if your strategy is valid, then some variant of this untested code: replicate (