Hi R-list, I have a data set with plot locations and observations and want to label them based on locations. For example, I have GPS information (x and y) as follows:> x[1] -87.85092 -87.85092 -87.85092 -87.85093 -87.85093 -87.85093 -87.85094 [8] -87.85094 -87.85094 -87.85096 -87.85095 -87.85095 -87.85095 -87.85096 [15] -87.85096 -87.85096 -87.85096 -87.85088 -87.85088 -87.85087 -87.85087 [22] -87.85087 -87.85087 -87.85086 -87.85086 -87.85086 -87.85085 -87.85086 [29] -87.85085 -87.85085 -87.85084 -87.85084 -87.85084 -87.85084 -87.85075 [36] -87.85075 -87.85076 -87.85076 -87.85077 -87.85076 -87.85076 -87.85076 [43] -87.85077 -87.85077 -87.85077 -87.85077 -87.85077 -87.85077 -87.85070 [50] -87.85072 -87.85073 -87.85075 -87.85078 -87.85079 -87.85082 -87.85084 [57] -87.85077 -87.85078 -87.85078 -87.85078 -87.85078 -87.85078 -87.85079 [64] -87.85079 -87.85080 -87.85080 -87.85071 -87.85071 -87.85071 -87.85070 [71] -87.85071 -87.85079 -87.85071 -87.85070 -87.85070 -87.85069 -87.85069 [78] -87.85069 -87.85069 -87.85068 -87.85068 -87.85068 -87.85067 -87.85059 [85] -87.85060 -87.85060 -87.85060 -87.85061 -87.85061 -87.85061 -87.85061 [92] -87.85061 -87.85062 -87.85062 -87.85062 -87.85062 -87.85063 -87.85063 [99] -87.85063 -87.85055 -87.85055 -87.85055 -87.85054 -87.85054 -87.85053 [106] -87.85053 -87.85053 -87.85053 -87.85053 -87.85052 -87.85052 -87.85052 [113] -87.85052 -87.85051 -87.85051 -87.85043 -87.85043 -87.85044 -87.85044 [120] -87.85044 -87.85045 -87.85045 -87.85045 -87.85045 -87.85046 -87.85046 [127] -87.85046 -87.85046 -87.85047 -87.85047 -87.85039 -87.85039 -87.85038 [134] -87.85038 -87.85038 -87.85037 -87.85037 -87.85037 -87.85037 -87.85036 [141] -87.85036 -87.85036 -87.85035 -87.85035 -87.85035 -87.85027 -87.85027 [148] -87.85027 -87.85027 -87.85028 -87.85028 -87.85028 -87.85029 -87.85029 [155] -87.85029 -87.85029 -87.85029 -87.85030 -87.85030 -87.85030 -87.85022 [162] -87.85022 -87.85022 -87.85021 -87.85021 -87.85021 -87.85020 -87.85020 [169] -87.85020 -87.85020 -87.85019 -87.85019 -87.85019 -87.85019 -87.85011 [176] -87.85011 -87.85011 -87.85011 -87.85012 -87.85012 -87.85012 -87.85012 [183] -87.85013 -87.85013 -87.85013 -87.85014 -87.85014 -87.85014 -87.85006 [190] -87.85006 -87.85006 -87.85005 -87.85005 -87.85004 -87.85004 -87.85004 [197] -87.85004 -87.85003 -87.85003 -87.85003 -87.85002 -87.85003 -87.84994 [204] -87.84994 -87.84995 -87.84995 -87.84995 -87.84995 -87.84996 -87.84996 [211] -87.84996 -87.84996 -87.84996 -87.84996 -87.84996 -87.84996 -87.84996 [218] -87.84996 -87.84996 -87.84996 -87.84996 -87.84990 -87.84991 -87.84993 [225] -87.84995 -87.84998 -87.84999 -87.85001 -87.85003 -87.84996 -87.84998 [232] -87.84997 -87.84998 -87.84989 -87.84990 -87.84989 -87.84989 -87.84988 [239] -87.84988 -87.84988 -87.84988 -87.84988 -87.84987 -87.84987 -87.84987 [246] -87.84987 -87.84978 -87.84978 -87.84979 -87.84979 -87.84979 -87.84979 [253] -87.84979 -87.84980 -87.84980 -87.84981 -87.84980 -87.84981 -87.84981 [260] -87.84973 -87.84973 -87.84973 -87.84972 -87.84972 -87.84972 -87.84971 [267] -87.84971 -87.84971 -87.84970 -87.84970 -87.84970 -87.84963 -87.84963 [274] -87.84963 -87.84963 -87.84963 -87.84964 -87.84964 -87.84965 -87.84964 [281] -87.84964 -87.84965 -87.84957 -87.84957 -87.84956 -87.84956 -87.84958 [288] -87.84958> y[1] 33.90342 33.90335 33.90328 33.90321 33.90314 33.90308 33.90301 33.90294 [9] 33.90287 33.90280 33.90274 33.90267 33.90260 33.90253 33.90246 33.90240 [17] 33.90233 33.90232 33.90239 33.90245 33.90252 33.90259 33.90266 33.90273 [25] 33.90279 33.90286 33.90293 33.90300 33.90307 33.90314 33.90321 33.90327 [33] 33.90334 33.90339 33.90337 33.90335 33.90328 33.90321 33.90319 33.90318 [41] 33.90317 33.90316 33.90315 33.90313 33.90312 33.90310 33.90309 33.90307 [49] 33.90314 33.90314 33.90314 33.90314 33.90314 33.90314 33.90314 33.90314 [57] 33.90300 33.90294 33.90287 33.90280 33.90273 33.90266 33.90252 33.90245 [65] 33.90239 33.90232 33.90231 33.90237 33.90245 33.90251 33.90258 33.90259 [73] 33.90265 33.90272 33.90279 33.90286 33.90292 33.90299 33.90306 33.90313 [81] 33.90320 33.90326 33.90334 33.90332 33.90327 33.90320 33.90314 33.90307 [89] 33.90300 33.90293 33.90286 33.90279 33.90272 33.90265 33.90258 33.90252 [97] 33.90245 33.90238 33.90231 33.90231 33.90237 33.90243 33.90250 33.90257 [105] 33.90264 33.90271 33.90278 33.90285 33.90292 33.90298 33.90306 33.90312 [113] 33.90319 33.90326 33.90329 33.90326 33.90319 33.90312 33.90306 33.90299 [121] 33.90292 33.90286 33.90279 33.90272 33.90265 33.90258 33.90251 33.90245 [129] 33.90237 33.90231 33.90230 33.90236 33.90243 33.90250 33.90257 33.90264 [137] 33.90271 33.90277 33.90284 33.90291 33.90298 33.90305 33.90311 33.90319 [145] 33.90325 33.90323 33.90319 33.90312 33.90305 33.90299 33.90291 33.90285 [153] 33.90278 33.90272 33.90264 33.90257 33.90250 33.90243 33.90237 33.90230 [161] 33.90229 33.90235 33.90243 33.90250 33.90256 33.90263 33.90270 33.90277 [169] 33.90283 33.90290 33.90297 33.90304 33.90311 33.90318 33.90318 33.90311 [177] 33.90304 33.90298 33.90291 33.90284 33.90277 33.90270 33.90263 33.90256 [185] 33.90250 33.90243 33.90236 33.90229 33.90229 33.90235 33.90242 33.90249 [193] 33.90256 33.90262 33.90269 33.90276 33.90283 33.90290 33.90297 33.90303 [201] 33.90311 33.90316 33.90315 33.90311 33.90304 33.90297 33.90290 33.90283 [209] 33.90270 33.90263 33.90262 33.90260 33.90259 33.90258 33.90257 33.90256 [217] 33.90255 33.90254 33.90253 33.90251 33.90250 33.90255 33.90255 33.90255 [225] 33.90256 33.90256 33.90256 33.90256 33.90256 33.90249 33.90243 33.90236 [233] 33.90229 33.90228 33.90235 33.90242 33.90248 33.90255 33.90262 33.90269 [241] 33.90276 33.90283 33.90289 33.90296 33.90304 33.90310 33.90310 33.90304 [249] 33.90297 33.90290 33.90284 33.90276 33.90269 33.90263 33.90256 33.90249 [257] 33.90242 33.90235 33.90228 33.90227 33.90234 33.90241 33.90247 33.90255 [265] 33.90261 33.90268 33.90275 33.90282 33.90289 33.90296 33.90303 33.90297 [273] 33.90289 33.90283 33.90275 33.90269 33.90262 33.90255 33.90248 33.90242 [281] 33.90235 33.90228 33.90227 33.90234 33.90241 33.90247 33.90254 33.90261 I mapped the locations of sampling plots based on x and y.> plot(x,y)I made arbitrarily 'fake' data with certain means and standard deviations to generate regression line as follows:> (xsim <- sort(rnorm(30, xmn, xsd)))[1] -87.85130 -87.85125 -87.85093 -87.85092 -87.85081 -87.85074 -87.85070 [8] -87.85070 -87.85060 -87.85059 -87.85053 -87.85046 -87.85046 -87.85042 [15] -87.85037 -87.85028 -87.85026 -87.85026 -87.85025 -87.85023 -87.85019 [22] -87.85015 -87.85007 -87.85003 -87.85000 -87.84970 -87.84957 -87.84957 [29] -87.84947 -87.84941> (ysim <- sort(rnorm(30, ymn, ysd)))[1] 33.90227 33.90237 33.90239 33.90240 33.90243 33.90248 33.90249 33.90251 [9] 33.90251 33.90262 33.90262 33.90266 33.90267 33.90268 33.90272 33.90274 [17] 33.90274 33.90278 33.90283 33.90286 33.90286 33.90294 33.90295 33.90301 [25] 33.90304 33.90316 33.90317 33.90318 33.90319 33.90321> (fm1 <- lm(ysim~xsim))Call: lm(formula = ysim ~ xsim) Coefficients: (Intercept) xsim 82.9996 0.5589 I overlapped fitted line on the plot.> abline(fm1)My question is: As you can see in the plot, how can I label (or re-group) those in upper diagonal as (say) 'A' and the others in lower diagonal as 'B'? Thanks a lot in advance!!! Steve [[alternative HTML version deleted]]
Hi Steve,
Just test whether y is greater than the predicted y (i.e., your line).
## function using the model coefficients*
f <- function(x) {82.9996 + (.5589 * x)}
## Find group membership
group <- ifelse(y > foo(x), "A", "B")
*Note that depending how accurate this needs to be, you will probably
want to use the model itself rather than just reading from the
printout like I did. If you need to do that, take a look at ?predict
For future reference, it would be easier for readers if you provided
your data via something like: dput(x) that can be copied directly into
the R console. Also, if you are generating random data (rnorm()), you
can use set.seed() so that we can replicate exactly what you get.
HTH,
Josh
On Fri, Mar 4, 2011 at 1:39 PM, Steve Hong <emptican at gmail.com>
wrote:> Hi R-list,
>
> I have a data set with plot locations and observations and want to label
> them based on locations. ?For example, I have GPS information (x and y) as
> follows:
[snip]>> (fm1 <- lm(ysim~xsim))
> Call:
> lm(formula = ysim ~ xsim)
> Coefficients:
> (Intercept) ? ? ? ? xsim
> ? ?82.9996 ? ? ? 0.5589
>
> I overlapped fitted line on the plot.
>
>> abline(fm1)
> My question is:
> As you can see in the plot, how can I label (or re-group) those in upper
> diagonal as (say) 'A' and the others in lower diagonal as
'B'?
>
> Thanks a lot in advance!!!
>
> Steve
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
--
Joshua Wiley
Ph.D. Student, Health Psychology
University of California, Los Angeles
http://www.joshuawiley.com/