Creating window intervals of the positions in R - r

I am creating an interval list of the positions, by window 4 and step 1.
pos <- subs$variable
intervals <- paste0(pos[seq(1, n, by=1)],":", pos[seq(4, n, by=1)])
intervals I get are:
[1] "92:107" "101:120" "106:132" "107:136" "120:140" "132:146" "136:147" "140:152" "146:166" "147:167" "152:174"
[12] "166:186" "167:187" "174:189" "186:204" "187:228" "189:229" "204:107" "228:120" "229:132"
subs <- structure(list(variable = c(92, 101, 106, 107, 120, 132, 136,
140, 146, 147, 152, 166, 167, 174, 186, 187, 189, 204, 228, 229
), covMean = c(11355.658, 11450.079, 11479.711, 11495.132, 11612.053,
11580.158, 11421.684, 11288.105, 11278, 11239.763, 11236.895,
10425.526, 10386.789, 10233.816, 9523.132, 9503.316, 9450.158,
8532.763, 7795.368, 7656.895), emboss = c(1.3717151, 1.3828546,
1.3880071, 1.3879077, 1.3959816, 1.3830276, 1.3724465, 1.363247,
1.337794, 1.3334625, 1.3097811, 1.2826296, 1.2811749, 1.2676601,
1.2193303, 1.2136416, 1.2044259, 1.1649542, 1.1163399, 1.1125204
)), row.names = c(56L, 62L, 65L, 67L, 77L, 82L, 87L, 95L, 97L,
114L, 119L, 133L, 154L, 156L, 169L, 173L, 186L, 190L, 212L, 214L
), class = "data.frame")
Why it starts iterating from the beginning, I want it to stop at the last position.


seq(1, n, by=1) and seq(4, n, by=1) are not of same length, the values are recycled. Try :
n <- length(pos)
intervals <- paste(pos[1:(n-3)],pos[4:n], sep =':')
intervals
# [1] "92:107" "101:120" "106:132" "107:136" "120:140" "132:146" "136:147" "140:152" "146:166"
#[10] "147:167" "152:174" "166:186" "167:187" "174:189" "186:204" "187:228" "189:229"

Related

How to plot specific columns in multiple dataframes in a list?

Sorry for this basic question. I thought I'd be able to solve this but I really could not figure out the syntax. I am quite new to R. Hopefully someone could explain to me how to do this.
I have a list of data frames with data and I wish to plot them.
df_list <- list(
`1.3.A` =
tibble::tribble(
~Person, ~Height, ~Weight,
"Alex", 175L, 75L,
"Gerard", 180L, 85L,
"Clyde", 179L, 79L
),
`2.2.A` =
tibble::tribble(
~Person, ~Height, ~Weight,
"Missy", 175L, 75L,
"Britany", 180L, 85L,
"Sussie", 179L, 79L
),
`1.1.B` =
tibble::tribble(
~Person, ~Height, ~Weight,
"Luke", 175L, 75L,
"Alex", 180L, 85L,
"Haley", 179L, 79L
)
)
Is there a way to plot the Height and Weight columns using ggdensity(). I am trying to make multiple density plots of each columns (Height and Weight) of each data frames but failed to do it.
Some codes I tried:
make_hist <- function(x){
ggdensity(data, x)
}
plots <- lapply(df_list, make_hist)
r <- lapply(df_list, function(x) {
ggdensity(data, x)
I even tried using imap but could not figure out the proper syntax for ggdensity and wrap it with imap() function or lapply().
Really sorry for this basic question and I hope someone could explain me the answer.

We can loop over the 'df_list', create the density plots for both 'Height' and 'Weight', then flatten the nested list (do.call(c) and use ggarrange to plot the elements in the list by specifying the nrow and ncol
library(ggpubr)
plots <- lapply(df_list, function(x) ggdensity(x, c("Height", "Weight")))
plots1 <- do.call(c, plots)
ggarrange(plotlist = plots1, nrow = 3, ncol = 2)
-output

How do I efficiently summarize summary output from multiple GAM models?

I am running multiple GAM models and need to view and compare the summary output from these. I'd like a quick and efficient way to extract and compile summary statistics from the models but have not found a way to do so.
A example data set is provided below:
example.data <- structure(list(response = c(1.47, 0.84, 1.99, 2.29, 4.14, 4.47,
2.71, 1.67, 4.12, 1.67, 2.03, 1.74, 0.98, 0.96, 0.56, 2.45, 1.31,
3.06, 2.35, 3.2, 1.16, 2.07, 0.99, 1.35, 1.02, 2.92, 1.8, 2.17,
2.56, 1.56, 2.33, 3.19, 1.53, 2.94, 3.28, 1.53, 2.8, 5.53, 1.26,
2.43, 3.5, 2.22, 3.73, 2.46, 2.16, 1.99, 3.34, 2.63, 2.51, 1.78
), predictor1 = c(17, 14.4, 99.45, 10.8, 54.25, 55.1, 40, 9,
54.25, 14.4, 14.4, 17, 14.4, 17, 10.8, 54.25, 54.25, 15.3, 55.1,
54.25, 14.4, 58, 17, 53.425, 58, 40.45, 14.4, 12.75, 91.05, 6.24,
100.25, 77.25, 43.4, 183.6, 91.05, 9.84, 100.25, 64, 10, 10,
91.05, 8.25, 100.25, 54.25, 89.4, 9.84, 10.8, 54.25, 10.8, 54.25
), predictor2 = c(165.7, 177.3, 594.2, 192.5, 426.2, 270.8, 244,
236.1, 416, 175.8, 258.6, 233.5, 115.8, 141, 153.5, 414.2, 438.9,
203, 261.4, 357.8, 148, 205.5, 137.4, 214.7, 167.8, 371.4, 179.9,
273.7, 567.2, 231.5, 355.3, 270, 319.5, 301.9, 301.9, 215.5,
256.5, 417, 231.8, 284.6, 396.3, 323, 458.4, 290, 203, 198, 350.8,
338, 323.5, 264.7), predictor3 = c(829.8, 841, 903.6, 870.3,
794, 745, 845.2, 906.5, 890.3, 874.2, 805.4, 828.8, 872, 854.7,
912.2, 790.8, 759.2, 855.1, 741.6, 961.8, 839.9, 805.1, 885.2,
887.8, 833.9, 1050.9, 787.5, 837, 731.9, 774.4, 820.8, 995.8,
916.3, 1032.1, 1014.3, 773.7, 846.4, 723.7, 764.2, 708.3, 1009.3,
1053.7, 751.7, 901.1, 848.7, 796.5, 697.1, 733.6, 725.6, 856.6
)), row.names = c(50L, 51L, 52L, 53L, 54L, 55L, 56L, 57L, 58L,
60L, 61L, 62L, 63L, 64L, 65L, 66L, 67L, 68L, 69L, 70L, 71L, 72L,
73L, 74L, 75L, 76L, 77L, 78L, 79L, 80L, 81L, 82L, 83L, 84L, 85L,
86L, 87L, 88L, 89L, 90L, 91L, 92L, 93L, 94L, 95L, 96L, 97L, 98L,
99L, 100L), class = "data.frame")
Right now, the unsophisticated and inefficient way I do it is something like this:
library(mgcv)
mod1 = gam(response ~ s(predictor1), data=example.data)
mod2 = gam(response ~ s(predictor2), data=example.data)
mod3 = gam(response ~ s(predictor3), data=example.data)
mod.names <- c("mod1", "mod2", "mod3")
mod.predictors <- c("predictor1", "predictor2", "predictor3")
mod.rsq <- c(summary(mod1)$r.sq, summary(mod2)$r.sq, summary(mod3)$r.sq)
mod.AIC <- c(AIC(mod1), AIC(mod2), AIC(mod3))
summary.data <- data.frame(mod.names,
mod.rsq,
mod.AIC,
mod.predictors)
summary.data
I can then select models accordingly from the summary table.
I have over one hundred potential predictors in the actual data, and it's obviously laborious to manually specify all the models and their output so a more automated alternative would be desirable.

The hard part of this question is choosing which models to run: that's a hard statistical question, and depending on what you choose, a less hard programming problem.
I'll assume that you are only interested in models like the ones in your example. Then this should work:
library(mgcv)
#> Loading required package: nlme
#> This is mgcv 1.8-33. For overview type 'help("mgcv-package")'.
predictors <- setdiff(names(example.data), "response")
result <- data.frame(predictors = predictors, rsq = NA, AIC = NA)
model <- response ~ predictor
for (i in seq_len(nrow(result))) {
pred <- result$predictors[i]
model[[3]] <- bquote(s(.(as.name(pred))))
mod <- gam(model, data = example.data)
result$rsq[i] <- summary(mod)$r.sq
result$AIC[i] <- AIC(mod)
}
result
#> predictors rsq AIC
#> 1 predictor1 0.2011252 138.0875
#> 2 predictor2 0.4666861 118.7270
#> 3 predictor3 0.1959123 139.0365
The tricky part is computing the model formula. I start with a simple model response ~ predictor, then replace the 3rd part (predictor) with code produced by bquote(s(.(as.name(pred)))). That function produces unevaluated code like s(predictor1) when pred holds "predictor1".

Creating and adding median, mean etc. in ggplot

I have a data frame R_m which looks like this:
data frame R_m
I have used pivot_longer to modify the data for ggplot and then print it:
R_m2 <- R_m %>%
pivot_longer(names_to = "per", values_to="ind", cols=-sim, names_ptypes=list(per=integer()))
ggplot(R_m2, aes(x=per, y=ind,color=sim, group=sim))+geom_line() +
theme(legend.position = "none")
Now, I would like to add a line for mean, median and some quantiles in the graph. Before, I got this data from elsewhere and imported it as a data frame 'stat' into R. It looks like this:
median, mean and quantile
which plotted nicely by adding
geom_line(data=subset(stat,sim=="Median"),colour="black", size = 1)
Outcome looks like this:
Now I would like to achieve the same in R. Creating an array of medians (where median is taken across each column in R_m) was straightforward to do. But then my first stumbling block was that using rbind, cbind etc. I could not create a data frame that looks like 'stat'. I could not rename the zero column to "sim", so the geom_line command above no longer works. Anyway, this requires a lot of data manipulation. Is there a more efficient way of adding a median, mean and percentile lines to a graph?
structure(list(sim = c(166, 37, 163, 65, 95, 92, 98, 168, 19,
157, 177, 200, 115, 177, 149, 130, 66, 114, 96, 12, 138, 39,
80, 33, 157, 107, 180, 159, 166, 14, 126, 67, 190, 86, 147, 182,
43, 5, 109, 141, 53, 186, 49, 68, 168, 107, 67, 28, 158, 178),
per = c(407L, 1763L, 2158L, 1608L, 836L, 1638L, 285L, 1978L,
45L, 1927L, 192L, 1517L, 163L, 789L, 1989L, 2478L, 2410L,
2445L, 1532L, 181L, 1489L, 1434L, 2515L, 676L, 1503L, 2458L,
732L, 1266L, 1705L, 1852L, 1543L, 1568L, 41L, 1992L, 600L,
1314L, 33L, 199L, 370L, 46L, 1171L, 1173L, 2048L, 994L, 836L,
372L, 2374L, 1414L, 1628L, 1188L), ind = c(97.7428137181456,
100.462039003802, 95.2793483563514, 98.3721036305918, 99.0584691732282,
103.301132288618, 102.428408453689, 100.387198613893, 99.7888039221544,
101.017059784079, 106.12288506898, 102.636429823681, 93.8144062244855,
104.280572544198, 97.3182467653953, 96.5603916025096, 96.3529141792467,
98.3149638711415, 98.3629878947972, 94.6106342501915, 99.6835722307572,
98.1716050345778, 103.055895201755, 100.054976695486, 96.1369802984859,
98.5257212288309, 98.8568719059079, 102.900859147552, 99.37215427561,
102.623437273663, 104.128600607447, 102.673062489082, 100.368131206055,
98.3487549118012, 96.4401682804699, 96.4407823981984, 97.9413312935541,
102.122624393907, 98.2979203190445, 101.018531709501, 100.444354410774,
101.118257199515, 100.867412455804, 98.9923953588876, 100.417977446024,
102.21423103019, 102.296794518966, 99.9367239162818, 102.314273028354,
100.80711113148)), row.names = c(NA, -50L), class = c("tbl_df",
"tbl", "data.frame"))

Remove steps in time series

I am analyzing lizards increase of temperature. Problem is that the animals sometime move moving the thermometer we are using for the measurements. This results in spikes (few measurements extremely high or low) and steps (sudden offsets of the curve). I wanted to clean my curves from this noise, removing the spikes and aligning the two parts of the curve separated by each step. I guess the spikes can be easily smoothed but what about the steps?
Examples of time series with steps marked in yellow and spikes in red
This is a sample from my dataframe
structure(list(ID = structure(c(183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L, 183L,
183L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L, 194L,
194L, 194L, 194L), .Label = c("101", "102", "104", "106", "107",
"109", "110", "111", "112", "113", "114", "115", "116", "118",
"119", "121", "122", "123", "124", "125", "126", "128", "129",
"130", "132", "133", "134", "157", "158", "161", "163", "164",
"165", "166", "167", "168", "169", "170", "171", "172", "173",
"174", "175", "177", "178", "179", "180", "181", "182", "183",
"186", "187", "188", "189", "191", "192", "193", "194", "195",
"196", "198", "199", "201", "202", "203", "204", "205", "207",
"208", "209", "210", "211", "213", "214", "215", "218", "219",
"220", "221", "222", "223", "224", "225", "226", "227", "228",
"229", "230", "233", "235", "307", "308", "310", "311", "312",
"313", "314", "315", "316", "317", "318", "319", "320", "321",
"322", "323", "324", "325", "326", "327", "328", "329", "330",
"331", "332", "333", "334", "335", "336", "339", "340", "341",
"343", "344", "346", "347", "348", "349", "350", "351", "352",
"353", "354", "355", "356", "357", "358", "58", "59", "60", "61",
"62", "63", "64", "65", "66", "67", "68", "69", "70", "71", "72",
"73", "74", "75", "76", "78", "79", "80", "81", "82", "83", "84",
"85", "86", "87", "88", "89", "90", "91", "92", "93", "94", "95",
"96", "97", "98", "99", "F04000017", "F04001300", "F04060000",
"F04070000", "F04080000", "F05000017", "F05000020", "F05060000",
"F05070000", "FSUM", "M03100000", "M04000016", "M04090000", "M04100000",
"M05000016", "M05000018", "M05080000", "M05090000", "M05100000",
"MSUM"), class = "factor"), TIME = c(1600, 0, 180, 280, 1270,
190, 1570, 100, 630, 110, 1530, 790, 90, 650, 910, 1520, 460,
270, 710, 1240, 1610, 550, 470, 1320, 360, 1220, 860, 540, 290,
1330, 1180, 170, 1700, 990, 1060, 420, 10, 330, 1110, 1160, 890,
260, 620, 1140, 1420, 380, 300, 1650, 1430, 200, 490, 1590, 150,
430, 720, 950, 730, 1660, 1260, 740, 810, 780, 1370, 960, 1130,
1210, 1670, 1010, 760, 1280, 1450, 770, 1070, 1000, 1030, 690,
80, 1040, 1050, 160, 1350, 1230, 1630, 640, 1390, 1460, 1100,
850, 1120, 390, 1170, 980, 320, 590, 520, 1090, 560, 870, 1020,
750, 140, 700, 70, 1500, 340, 1290, 530, 840, 230, 370, 1440,
1200, 1480, 30, 1400, 210, 1300, 480, 450, 1580, 1560, 410, 1360,
900, 670, 1540, 50, 500, 400, 1380, 930, 580, 1680, 600, 1340,
240, 1190, 570, 1640, 940, 350, 1550, 830, 310, 880, 1080, 920,
1490, 610, 220, 1470, 800, 820, 40, 1250, 130, 60, 680, 20, 970,
1620, 440, 1690, 120, 510, 1410, 660, 250, 1310, 1150, 1510,
1230, 1540, 320, 1380, 680, 410, 480, 580, 130, 670, 1200, 1900,
1410, 1470, 1950, 400, 870, 970, 1740, 500, 1300, 1220, 240,
660, 1100, 20, 490, 590, 900, 820, 1910, 920, 690, 1290, 1710,
750, 1860, 1840, 880, 1250, 2070, 640, 1580, 1310, 800, 1000,
2090, 1760, 330, 760, 40, 600, 1800, 200, 1730, 2080, 470, 30,
1490, 2030, 860, 1620, 250, 1880, 790, 90, 360, 1520, 460, 180,
1770, 1450, 1940, 1370, 1180, 290, 2060, 1240, 1500, 1480, 510,
980, 1610, 1630, 950, 1980, 990, 140, 60, 110, 1680, 10, 550,
1700, 1750, 1270, 1690, 230, 2010, 1070, 910, 300, 1850, 1140,
1350, 930, 170, 1650, 1170, 1820, 1040, 1420, 120, 1150, 50,
960, 1560, 1790, 630, 280, 740, 1890, 2020, 370, 2040, 2050,
310, 380, 1920, 340, 1030, 430, 1330, 1670, 1960, 830, 1590,
420, 1400, 560, 1360, 1460, 1440, 1120, 1870, 1720, 1130, 730,
390, 850, 520, 1090, 1550, 100, 1930, 1010, 440, 210, 1390, 720,
1190, 450, 770, 1320, 220, 1640, 570, 2100, 1210, 650, 2000,
840, 1340, 1830, 530, 1110, 1260, 890, 700, 80, 810, 1060, 260,
1510, 1600, 70, 1430, 1280, 1530, 1020, 540, 940, 1050, 710,
1660, 1570, 270, 610, 620, 780, 1780, 350, 1970, 0, 1160, 1990,
190, 150, 1080, 160, 1810), LIZ = c(33.88628, 16.01848, 19.07537,
20.7521, 31.20864, 19.28933, 33.81645, 17.74926, 25.17476, 17.96775,
35.05947, 27.00651, 17.59508, 25.4069, 28.208, 34.99299, 22.88728,
20.59275, 26.18355, 31.03587, 33.947, 24.10997, 23.02703, 33.7095,
22.17608, 30.91595, 27.75973, 23.97134, 20.93754, 33.80886, 30.68742,
18.91375, 33.92175, 29.09865, 29.63962, 22.70535, 16.21696, 21.63548,
30.08059, 30.53127, 27.98175, 20.43409, 25.05877, 30.36036, 34.46184,
22.44901, 21.10243, 33.94136, 34.52333, 19.46398, 23.28305, 33.80731,
18.61317, 22.47734, 26.25653, 28.72091, 26.38854, 33.93099, 31.15667,
26.51461, 27.30298, 26.90585, 34.10386, 28.83213, 30.1994, 30.83049,
33.89071, 29.26265, 26.67329, 32.40561, 34.64661, 26.80018, 29.67405,
29.16813, 29.42029, 25.87777, 17.42933, 29.50874, 29.60253, 18.75925,
33.97794, 30.99272, 33.94607, 25.2893, 34.24567, 34.68945, 30.0688,
27.65221, 30.13411, 22.59461, 30.63778, 28.96221, 21.41397, 24.58403,
23.68083, 29.85481, 24.22983, 27.8342, 29.36583, 26.60143, 18.46704,
26.03891, 17.26645, 34.9014, 21.84469, 33.39574, 23.82, 27.57049,
19.93821, 22.30472, 34.60674, 30.77051, 34.79948, 16.55761, 34.33158,
19.62092, 33.46234, 23.15645, 22.68871, 33.79304, 33.88377, 22.99619,
34.04042, 28.1129, 25.6251, 34.21122, 16.89074, 23.43516, 22.78885,
34.17764, 28.47507, 24.45472, 33.8742, 24.72735, 33.90453, 20.10577,
30.73417, 24.35771, 33.93048, 28.62741, 21.99781, 33.97049, 27.48719,
21.25535, 27.92438, 29.80396, 28.35731, 34.8615, 24.93956, 19.783,
34.76926, 27.19699, 27.40897, 16.72252, 31.08047, 18.32881, 17.0416,
25.72955, 16.38935, 28.8926, 33.97636, 22.53829, 33.86432, 18.22302,
23.55972, 34.38583, 25.50873, 20.27311, 33.6045, 30.43855, 34.93784,
35.288345, 36.257075, 25.391195, 36.09727, 30.813695, 27.044815,
28.04567, 29.548165, 20.606055, 30.69551, 35.129335, 37.32639,
36.37374, 36.308585, 37.41784, 26.86707, 32.548155, 33.36402,
37.37834, 28.385865, 35.88449, 35.231535, 23.53661, 30.580155,
34.58997, 16.9712, 28.21824, 29.69837, 32.803, 32.16565, 37.3345,
32.96087, 30.917135, 35.835545, 37.40881, 31.57274, 37.28267,
37.3545, 32.635805, 35.639945, 37.55586, 30.322245, 36.361875,
35.93363, 32.083515, 33.9122, 37.57389, 37.38447, 25.60093, 31.70867,
17.74926, 29.83628, 37.32308, 22.486975, 37.37677, 37.54575,
27.8517, 17.37036, 36.13542, 37.58466, 32.4728, 36.84912, 23.79395,
37.2999, 32.03574, 19.402135, 27.835535, 36.185425, 27.68629,
21.91643, 37.37574, 36.612895, 37.38406, 36.11211, 35.03594,
24.74032, 37.55308, 35.4775, 36.152685, 36.125095, 28.51215,
33.614975, 36.71473, 36.859965, 33.190425, 37.57786, 33.82105,
20.911715, 18.444165, 19.9942, 36.89602, 16.551, 29.161815, 37.4295,
37.38779, 35.75298, 36.91141, 23.29514, 37.59916, 34.393665,
32.88079, 24.885135, 37.30579, 34.812115, 36.05718, 33.03272,
21.730805, 36.877595, 34.980885, 37.34123, 34.19463, 36.436875,
20.27451, 34.880975, 18.10635, 33.268755, 36.24434, 37.33012,
30.18474, 24.530635, 31.445215, 37.32058, 37.6003, 36.417575,
37.57818, 37.56227, 25.15269, 26.548085, 37.34238, 25.799625,
34.12894, 27.14156, 36.0024, 36.89349, 37.48383, 32.23698, 36.386895,
27.013425, 36.295295, 29.29934, 36.091595, 36.642735, 36.5601,
34.706135, 37.27491, 37.38146, 34.758915, 31.35675, 26.709435,
32.374665, 28.69313, 34.52595, 36.22561, 19.717425, 37.35049,
33.991875, 27.288855, 22.76699, 36.179625, 31.25419, 35.08752,
27.507965, 31.821405, 35.97357, 23.037415, 36.86924, 29.408195,
37.60048, 35.18184, 30.45653, 37.58567, 32.27861, 36.04241, 37.34733,
28.86293, 34.64831, 35.684655, 32.71743, 31.01652, 19.07221,
32.1161, 34.32663, 24.059185, 36.16283, 36.413085, 18.77961,
36.49902, 35.7932, 36.214135, 34.05778, 29.01832, 33.111965,
34.2606, 31.14351, 36.88914, 36.326515, 24.30303, 29.97399, 30.103995,
31.92968, 37.36615, 26.01068, 37.52906, 16.12443, 34.929045,
37.59149, 22.175545, 21.19651, 34.4613, 21.473975, 37.33004),
COP = c(22.39478, 22.68049, 22.27224, 22.19685, 22.16573,
22.26254, 22.34891, 22.41931, 22.15782, 22.39973, 22.31492,
22.13082, 22.44822, 22.16036, 22.13963, 22.3111, 22.15529,
22.20868, 22.14849, 22.16546, 22.40751, 22.15873, 22.15807,
22.15667, 22.17006, 22.16576, 22.13619, 22.15633, 22.18792,
22.15968, 22.15784, 22.28282, 22.39207, 22.15544, 22.15307,
22.15378, 22.65061, 22.18484, 22.14693, 22.1517, 22.14945,
22.22223, 22.15782, 22.15069, 22.18498, 22.17095, 22.18882,
22.44182, 22.18867, 22.25752, 22.16363, 22.37156, 22.33576,
22.14907, 22.15658, 22.15133, 22.1547, 22.43926, 22.16738,
22.15422, 22.13264, 22.14114, 22.17317, 22.15817, 22.14417,
22.16643, 22.43323, 22.14401, 22.14921, 22.15957, 22.22265,
22.14089, 22.14463, 22.14593, 22.14691, 22.14661, 22.47322,
22.15314, 22.15694, 22.3083, 22.16482, 22.16823, 22.42285,
22.16284, 22.17225, 22.2391, 22.1465, 22.12234, 22.14539,
22.15492, 22.14823, 22.15392, 22.18663, 22.14773, 22.16047,
22.14056, 22.15791, 22.14152, 22.14251, 22.1509, 22.35001,
22.14744, 22.49713, 22.28185, 22.18089, 22.15846, 22.15929,
22.12327, 22.23335, 22.16951, 22.20456, 22.16193, 22.25876,
22.60373, 22.17513, 22.25065, 22.15608, 22.16255, 22.14552,
22.36192, 22.3395, 22.15159, 22.16513, 22.14421, 22.15804,
22.31474, 22.5485, 22.16658, 22.15816, 22.16993, 22.14259,
22.14321, 22.4235, 22.16097, 22.16565, 22.22893, 22.16082,
22.14911, 22.43463, 22.14959, 22.16988, 22.32926, 22.12439,
22.18906, 22.14787, 22.143, 22.14448, 22.26607, 22.16002,
22.24653, 22.25201, 22.13589, 22.13117, 22.578, 22.16947,
22.36273, 22.51722, 22.15239, 22.62479, 22.15932, 22.41646,
22.14778, 22.41367, 22.38414, 22.16356, 22.18521, 22.16367,
22.22729, 22.16007, 22.15065, 22.29892, 21.72296, 21.62249,
22.017515, 21.693785, 21.70153, 21.87866, 21.81236, 21.732425,
22.36322, 21.70375, 21.72846, 21.91553, 21.68792, 21.660505,
22.00908, 21.89052, 21.763825, 21.760305, 21.83179, 21.79684,
21.71059, 21.726415, 22.149185, 21.70785, 21.75106, 22.571055,
21.804205, 21.7291, 21.761135, 21.747215, 21.92869, 21.76272,
21.705115, 21.712995, 21.83027, 21.708885, 21.83844, 21.84054,
21.768415, 21.71803, 22.11175, 21.70465, 21.60251, 21.70955,
21.742635, 21.753505, 22.13507, 21.82251, 21.999255, 21.717505,
22.53302, 21.72247, 21.82229, 22.22044, 21.829, 22.12267,
21.82473, 22.55087, 21.6558, 22.09163, 21.76516, 21.593285,
22.13412, 21.86871, 21.73596, 22.43231, 21.943965, 21.636055,
21.836155, 22.26584, 21.82428, 21.672995, 21.98196, 21.694165,
21.73181, 22.066775, 22.10273, 21.71976, 21.64993, 21.65895,
21.788935, 21.759365, 21.59079, 21.59272, 21.762905, 22.04811,
21.755645, 22.34494, 22.4932, 22.397445, 21.59399, 22.582895,
21.750065, 21.84207, 21.82297, 21.712635, 21.593685, 22.164485,
22.07698, 21.7517, 21.760385, 22.04753, 21.8382, 21.741645,
21.69029, 21.759415, 22.28582, 21.590015, 21.73355, 21.83193,
21.75102, 21.68036, 22.378585, 21.73784, 22.51333, 21.75792,
21.609515, 21.82392, 21.70592, 22.084645, 21.708, 21.89372,
22.08329, 21.93241, 22.10217, 22.10085, 22.030495, 21.91551,
21.94469, 21.979055, 21.75095, 21.858405, 21.698245, 21.59609,
22.02914, 21.755635, 21.593795, 21.86841, 21.689295, 21.739825,
21.691495, 21.66596, 21.674615, 21.74691, 21.84458, 21.82984,
21.741525, 21.710355, 21.90157, 21.762305, 21.77891, 21.750345,
21.61608, 22.41471, 21.95989, 21.7551, 21.84604, 22.198325,
21.691115, 21.709805, 21.732425, 21.84097, 21.723055, 21.70937,
22.179165, 21.58997, 21.732705, 22.13691, 21.725925, 21.70574,
22.06154, 21.755495, 21.692555, 21.83686, 21.77328, 21.74777,
21.716135, 21.761175, 21.709325, 22.451175, 21.749225, 21.750545,
22.117925, 21.636325, 21.59097, 22.47463, 21.67836, 21.715015,
21.63479, 21.75122, 21.762415, 21.764115, 21.751305, 21.70731,
21.591635, 21.606505, 22.09898, 21.71826, 21.71403, 21.727555,
21.81842, 21.963065, 22.03673, 22.597065, 21.73344, 22.05635,
22.243885, 22.325185, 21.75266, 22.3091, 21.82619), BLK = c(28.64989,
15.80412, 20.22131, 22.01231, 28.0591, 20.41488, 28.57393,
18.42224, 25.81396, 18.66597, 28.51571, 26.66224, 18.17439,
25.95201, 27.20529, 28.51703, 24.38048, 21.85311, 26.30937,
28.02257, 28.6773, 25.22704, 24.48749, 28.1236, 23.18583,
27.98485, 27.00689, 25.14024, 22.16927, 28.1335, 27.89662,
20.00687, 28.67155, 27.44175, 27.62627, 23.92489, 16.09932,
22.78544, 27.75227, 27.83675, 27.13084, 21.69284, 25.74085,
27.80579, 28.28594, 23.44893, 22.33534, 28.7166, 28.29154,
20.61311, 24.68904, 28.60215, 19.59335, 24.03528, 26.36296,
27.34839, 26.40381, 28.7205, 28.04637, 26.45016, 26.76595,
26.62852, 28.18898, 27.38003, 27.79384, 27.96629, 28.71652,
27.47686, 26.54292, 28.06953, 28.37088, 26.5854, 27.64778,
27.45691, 27.54817, 26.17697, 17.91728, 27.57453, 27.59667,
19.79719, 28.15391, 28.01052, 28.70145, 25.8918, 28.19863,
28.40318, 27.73245, 26.95575, 27.77656, 23.56018, 27.86381,
27.42115, 22.64232, 25.52273, 24.96468, 27.69186, 25.29434,
27.05567, 27.51177, 26.4991, 19.37665, 26.24547, 17.66023,
28.45211, 22.92771, 28.08868, 25.05819, 26.9147, 21.16986,
23.31623, 28.31888, 27.94109, 28.41742, 16.63277, 28.21869,
20.81001, 28.10906, 24.58388, 24.26295, 28.59208, 28.56833,
23.80437, 28.16685, 27.16407, 26.06582, 28.53314, 17.15336,
24.78023, 23.69012, 28.19161, 27.27431, 25.4426, 28.70709,
25.60821, 28.15239, 21.35028, 27.92147, 25.36057, 28.71294,
27.31714, 23.05455, 28.55633, 26.8609, 22.49237, 27.09901,
27.67287, 27.23569, 28.43207, 25.6761, 20.99499, 28.41302,
26.72662, 26.81219, 16.89905, 28.04126, 19.14027, 17.39888,
26.11992, 16.37336, 27.39998, 28.69902, 24.15099, 28.69344,
18.90929, 24.86793, 28.25387, 26.01419, 21.52158, 28.12068,
27.81882, 28.48342, 27.96163, 28.197645, 22.70333, 28.117945,
26.073955, 23.85303, 24.593665, 25.413525, 19.11083, 26.00622,
27.90615, 27.98023, 28.14475, 28.161525, 28.08333, 23.7368,
27.01432, 27.341995, 27.83233, 24.77127, 28.08601, 27.96112,
21.39113, 25.949465, 27.68429, 15.858655, 24.68182, 25.48676,
27.12372, 26.823595, 27.98649, 27.17823, 26.13994, 28.074105,
27.80747, 26.44518, 27.88772, 27.88865, 27.054655, 27.983055,
28.14709, 25.80761, 28.22729, 28.087765, 26.741425, 27.415515,
28.16615, 27.82628, 22.852945, 26.515965, 16.55507, 25.558165,
27.86497, 20.59904, 27.82338, 28.15537, 24.503635, 16.208215,
28.18434, 28.12301, 26.98152, 28.238175, 21.57631, 27.93972,
26.69019, 18.11697, 23.249475, 28.194915, 24.400325, 20.18436,
27.83363, 28.152575, 28.05914, 28.1046, 27.87944, 22.246695,
28.13895, 27.97113, 28.17759, 28.17134, 24.852555, 27.36403,
28.231645, 28.240475, 27.28467, 28.09842, 27.38207, 19.33762,
17.208895, 18.63795, 28.32022, 15.478795, 25.178505, 27.81531,
27.83335, 28.025865, 28.32901, 21.198015, 28.12642, 27.594495,
27.149095, 22.40141, 27.87721, 27.77537, 28.07577, 27.214415,
19.976595, 28.24768, 27.87019, 27.87417, 27.512335, 28.1487,
18.87582, 27.808485, 16.88221, 27.304755, 28.22269, 27.85485,
25.751105, 22.085525, 26.402235, 27.96385, 28.12362, 23.3511,
28.13134, 28.13854, 22.55042, 23.483435, 27.99603, 22.988955,
27.479825, 24.074565, 28.06897, 28.300515, 28.08899, 26.850145,
28.21175, 23.961985, 28.148655, 25.25804, 28.084975, 28.15017,
28.142015, 27.74106, 27.89856, 27.81585, 27.762185, 26.34704,
23.61241, 26.9402, 24.945815, 27.660965, 28.20952, 18.388425,
28.03487, 27.44823, 24.182, 20.80328, 28.138395, 26.294595,
27.8952, 24.294155, 26.569235, 28.076645, 21.003065, 28.221055,
25.336215, 28.17753, 27.934355, 25.886285, 28.10892, 26.890965,
28.06394, 27.89246, 25.030615, 27.71521, 27.994955, 27.091235,
26.189505, 17.829845, 26.78325, 27.566845, 21.749355, 28.181235,
28.216485, 17.523135, 28.145175, 28.05206, 28.198345, 27.46945,
25.099975, 27.266245, 27.54595, 26.236675, 28.27499, 28.230255,
21.91532, 25.631615, 25.69132, 26.62857, 27.83423, 23.119855,
28.08352, 15.10413, 27.84174, 28.10417, 20.39175, 19.55529,
27.62871, 19.76947, 27.87254)), row.names = c(39410L, 39411L,
39412L, 39413L, 39414L, 39415L, 39416L, 39417L, 39418L, 39419L,
39420L, 39421L, 39422L, 39423L, 39424L, 39425L, 39426L, 39427L,
39428L, 39429L, 39430L, 39431L, 39432L, 39433L, 39434L, 39435L,
39436L, 39437L, 39438L, 39439L, 39440L, 39441L, 39442L, 39443L,
39444L, 39445L, 39446L, 39447L, 39448L, 39449L, 39450L, 39451L,
39452L, 39453L, 39454L, 39455L, 39456L, 39457L, 39458L, 39459L,
39460L, 39461L, 39462L, 39463L, 39464L, 39465L, 39466L, 39467L,
39468L, 39469L, 39470L, 39471L, 39472L, 39473L, 39474L, 39475L,
39476L, 39477L, 39478L, 39479L, 39480L, 39481L, 39482L, 39483L,
39484L, 39485L, 39486L, 39487L, 39488L, 39489L, 39490L, 39491L,
39492L, 39493L, 39494L, 39495L, 39496L, 39497L, 39498L, 39499L,
39500L, 39501L, 39502L, 39503L, 39504L, 39505L, 39506L, 39507L,
39508L, 39509L, 39510L, 39511L, 39512L, 39513L, 39514L, 39515L,
39516L, 39517L, 39518L, 39519L, 39520L, 39521L, 39522L, 39523L,
39524L, 39525L, 39526L, 39527L, 39528L, 39529L, 39530L, 39531L,
39532L, 39533L, 39534L, 39535L, 39536L, 39537L, 39538L, 39539L,
39540L, 39541L, 39542L, 39543L, 39544L, 39545L, 39546L, 39547L,
39548L, 39549L, 39550L, 39551L, 39552L, 39553L, 39554L, 39555L,
39556L, 39557L, 39558L, 39559L, 39560L, 39561L, 39562L, 39563L,
39564L, 39565L, 39566L, 39567L, 39568L, 39569L, 39570L, 39571L,
39572L, 39573L, 39574L, 39575L, 39576L, 39577L, 39578L, 39579L,
39580L, 41926L, 41927L, 41928L, 41929L, 41930L, 41931L, 41932L,
41933L, 41934L, 41935L, 41936L, 41937L, 41938L, 41939L, 41940L,
41941L, 41942L, 41943L, 41944L, 41945L, 41946L, 41947L, 41948L,
41949L, 41950L, 41951L, 41952L, 41953L, 41954L, 41955L, 41956L,
41957L, 41958L, 41959L, 41960L, 41961L, 41962L, 41963L, 41964L,
41965L, 41966L, 41967L, 41968L, 41969L, 41970L, 41971L, 41972L,
41973L, 41974L, 41975L, 41976L, 41977L, 41978L, 41979L, 41980L,
41981L, 41982L, 41983L, 41984L, 41985L, 41986L, 41987L, 41988L,
41989L, 41990L, 41991L, 41992L, 41993L, 41994L, 41995L, 41996L,
41997L, 41998L, 41999L, 42000L, 42001L, 42002L, 42003L, 42004L,
42005L, 42006L, 42007L, 42008L, 42009L, 42010L, 42011L, 42012L,
42013L, 42014L, 42015L, 42016L, 42017L, 42018L, 42019L, 42020L,
42021L, 42022L, 42023L, 42024L, 42025L, 42026L, 42027L, 42028L,
42029L, 42030L, 42031L, 42032L, 42033L, 42034L, 42035L, 42036L,
42037L, 42038L, 42039L, 42040L, 42041L, 42042L, 42043L, 42044L,
42045L, 42046L, 42047L, 42048L, 42049L, 42050L, 42051L, 42052L,
42053L, 42054L, 42055L, 42056L, 42057L, 42058L, 42059L, 42060L,
42061L, 42062L, 42063L, 42064L, 42065L, 42066L, 42067L, 42068L,
42069L, 42070L, 42071L, 42072L, 42073L, 42074L, 42075L, 42076L,
42077L, 42078L, 42079L, 42080L, 42081L, 42082L, 42083L, 42084L,
42085L, 42086L, 42087L, 42088L, 42089L, 42090L, 42091L, 42092L,
42093L, 42094L, 42095L, 42096L, 42097L, 42098L, 42099L, 42100L,
42101L, 42102L, 42103L, 42104L, 42105L, 42106L, 42107L, 42108L,
42109L, 42110L, 42111L, 42112L, 42113L, 42114L, 42115L, 42116L,
42117L, 42118L, 42119L, 42120L, 42121L, 42122L, 42123L, 42124L,
42125L, 42126L, 42127L, 42128L, 42129L, 42130L, 42131L, 42132L,
42133L, 42134L, 42135L, 42136L), class = "data.frame")
My problem is actually very similar to the one that this person had in python
Remove jumps like peaks and steps in timeseries but I haven't been able to find something similar for R
EDIT: I actually found something similar R-related in this question https://stats.stackexchange.com/questions/139660/detecting-changes-in-time-series-r-example?newreg=f119230044de4802a9f0f6f4e4637d8f
The solution using tsoutliers looks applicable to my problem but so far it didn't work very well. I am not sure about what I am doing wrong.
tso(dat.ts, types = c("LS","TC"))
flattens completely my curve
tso(dat.ts, types = c("LS","TC"), discard.method = "bottom-up")
Moves my curve in the right direction but not enough to fix the problem.
Any approach using the bottom-up method to discard outliers gives this result, any approac using the en-masse approach flattens the curve.

There are a couple methods attempted in the code below attempting to capture the changes: by percent change and by using a rolling median method.
First import a couple useful libraries. Also changing your data into a tibble called 'lizard_data'
library(tidyverse)
library(RcppRoll)
lizard_data <- tibble(your_data)
Then using ggplot2, we can visualize all the time series.
lizard_data %>%
pivot_longer(names_to = 'key', values_to = 'value', cols = c(LIZ, COP, BLK)) %>%
arrange(TIME) %>%
ggplot(aes(x = TIME, y = value, color = key))+
geom_line()+
facet_wrap(~ID, nrow = 2)
Then we remove the adjustments by finding the delta, and we can use two different methods to smooth the time series. First, with a percentile method (changing the top 5% and bottom 5% of values to the median) and second, with the rolling median (if the max change is ten percent higher or lower than the median, replacing the value with the median).
ld_w_change <- lizard_data %>%
pivot_longer(names_to = 'key', values_to = 'value', cols = c(LIZ, COP, BLK)) %>%
group_by(ID, key) %>%
arrange(TIME) %>%
mutate(lag = lag(value),
raw_change = (value-lag),
#using the percentile change method
med_raw_change = median(raw_change, na.rm = T),
q_05 = quantile(raw_change,.05, na.rm = T),
q_95 = quantile(raw_change,.95, na.rm = T),
adj_raw_change = if_else(raw_change > q_95 | raw_change < q_05, med_raw_change, raw_change),
normalized_change = if_else(is.na(adj_raw_change), 0,adj_raw_change),
initial_value = first(value),
roll_raw_change = cumsum(normalized_change),
new_value_pct = initial_value + roll_raw_change,
# using the rolling median method
rolling_median_change = roll_median(raw_change, n = 5, align = "right", na.rm = T, fill = NA),
adj_median_change = case_when(raw_change > rolling_median_change*1.1 | raw_change < rolling_median_change*.9 ~ rolling_median_change,
is.na(rolling_median_change) & is.na(raw_change) ~ 0,
T ~ raw_change),
normalized_med_change = cumsum(adj_median_change),
new_value_roll = initial_value + normalized_med_change
)
I personally prefer the median adjustment, it seems to preserve the shape of the data better than the percentile method. To compare these methods visually, we can plot them side-by-side:
ld_w_change %>%
pivot_longer(names_to = 'method', values_to = 'adjusted_temp', cols = c(new_value_pct, new_value_roll)) %>%
ggplot(aes(x = TIME, y =adjusted_temp, color = key))+
geom_line()+
facet_wrap(ID ~ method, nrow = 2)

Plot slope and intercept of a model with log and percentage transformations

This is the structure of my data
> dput(test)
structure(list(MAT = c(4.9, 4.9, 15.5, 14.1, 14.1, 14.1, 11.5,
11.5, 11.5, 17, 6.1, 2.7, 2.2, 2.2, 14.1, 14.1, 14.1, 9.5, 9.5,
9.5, 9.5, 9.3, 8.3, 8.266666651, 8.266666651, 4.3, 4.3, 22.3,
14.1, 14.1, 14.1, 8.5, 8.5, 8.5, 8.5, 21.5, 21.5, 3.8, 3.8, 6,
6, 6, 6, 6), es = c(0.29603085763985, 0.421393627439682, 0.189653473156549,
0.226685054608428, 0.291373762079697, 0.166533544378467, 0.250586529054368,
0.146320008054403, 0.199565119644333, -0.0819047677231083, 0.15963948187092,
-0.154628141843561, 0.201121044198443, 0.0867981239977565, 0.543870310978598,
0.34547921143505, 0.37557241352574, -0.287318919407836, 0.207937483228907, 0.190143660810163, 0.276182673435993, 0.128596803172119, 0.454753165843559,
0.399237234440439, 0.32075358541748, 0.362664873575803, -0.0865925288159671,
0.51290512543514, 0.186308318839249, 0.147936083867325, 0.243792477087184,
0.625169403695832, 0.110317782120045, 0.217836235313289, 0.171468156841181,
0.50548821117127, 0.164418265301427, -0.00246305543239786, 0.325552346507191,
0.381240606108843, 0.19337350462531, 0.0408803528990759, 0.321815078821239,
0.307642815014319), var = c(0.00496277337027962, 0.0130962311273343,
0.0180149624217804, 0.0134568083459063, 0.00139708925143695,
0.000725862546533828, 0.00670831011660164, 0.0190783110089115,
0.0641568910090007, 0.0121596544795352, 0.0653909966557582, 0.0514610437228611,
0.0231592619167496, 0.0108989891148006, 0.0588577146414195, 0.0695760532112402,
0.0744256820906048, 0.00997789089155498, 0.00928124381998638,
0.0145009450673482, 0.00652956018299188, 0.0111886178917916,
0.0265943757419349, 0.142676904340634, 0.110705177803624, 0.0576538348777718,
0.0625171635976251, 0.0131652117394448, 0.00947904166717649,
0.00813569411386797, 0.00444289889858652, 0.0673007030900184,0.00545169559098343, 0.240046081413733, 0.00561125010476281,
0.0185516235174018, 0.0179989506841957, 0.0496806959944248, 0.022478393723115,
0.0521209786580004, 0.282298667080106, 0.0151428845076692, 0.00992945920656693, 0.0145544965304081), MAP = c(810, 810, 1140, 1750, 1750, 1750,
1034, 1034, 1034, 720, 645, 645, 645, 645, 1000, 1000, 1000,
691, 691, 691, 691, 1134, 1750, 1326, 1326, 1140, 1140, 1310,
1750, 1750, 1750, 1003, 1003, 1003, 1003, 1750, 1750, 1750, 1750,
1750, 1750, 1750, 1750, 1750), CO2dif = c(162L, 162L, 190L, 165L,
165L, 165L, 200L, 200L, 200L, 150L, 335L, 335L, 335L, 335L, 348L,
348L, 348L, 200L, 200L, 200L, 200L, 220L, 350L, 350L, 350L, 350L,
350L, 350L, 180L, 180L, 180L, 130L, 130L, 130L, 130L, 320L, 320L,
360L, 360L, 345L, 345L, 350L, 348L, 348L)), row.names = c(NA,
-44L), class = "data.frame", .Names = c("MAT", "es", "var", "MAP",
"CO2dif"))
I run model selection using meta-analysis, and the best model to predict the effects size is:
library(metafor)
summary(rma(es, var, data=test ,control=list(stepadj=.5), mods= ~ 1 + log(MAT) + MAP + CO2dif + log(MAT):CO2dif, knha=TRUE))
Model Results:
estimate se tval pval ci.lb ci.ub
intrcpt 1.2556 0.3719 3.3758 0.0017 0.5033 2.0080 **
log(MAT) -0.5740 0.1694 -3.3882 0.0016 -0.9167 -0.2313 **
MAP 0.0001 0.0001 2.5181 0.0160 0.0000 0.0003 *
CO2dif -0.0042 0.0013 -3.2932 0.0021 -0.0067 -0.0016 **
log(MAT):CO2dif 0.0020 0.0005 3.7500 0.0006 0.0009 0.0031 ***
Now I want to plot es vs MAT, with an example with this model, assuming that MAP=1200 mm and CO2dif=350
MAPi <- 1200
CO2i <- 350
make_pct <- function(x) (exp(x) - 1) * 100
ggplot(test, aes(x = log(MAT), y = make_pct(es))) +
geom_abline(aes(intercept = make_pct(1.2556 + 0.0001 * MAPi - 0.0042 * CO2i),
slope = make_pct(log(0.0020 * CO2i)) - make_pct(log(0.5740))) ,
color = "red", size=0.8) +
geom_point() +
theme_classic()
Effect size (es) is in log format, and I want percentage, so I transform it with the function make_pct. MAT, on the other hand, has to be log-transformed in the plot as indicated in the model output. Is the slope of the ggplot above correct with the log and percentage transformations? It seems to me that the slope is rather low. I am not very familiar with this type of plots and transformations, so any tips are welcome. Thanks

The relationship between exp(es)-1 and the explanatory variable log(MAT) is not linear.
For a given set of values of MAP and CO2dif, this relationship is of the form: y = exp(es)-1 = k1*exp(k2*log(MAT)).
This function can be plotted as follows:
library(metafor)
library(ggplot2)
modfit <- rma(es, var, data=test ,control=list(stepadj=.5),
mods= ~ 1 + MAP + log(MAT)*CO2dif, knha=TRUE)
pars <- coef(modfit)
MAPi <- 1200
CO2i <- 350
make_pct <- function(x) (exp(x) - 1) * 100
mod_fun <- function(MAP, MAT, CO2dif, pars) {
y <- pars[1]+pars[2]*MAP+pars[3]*log(MAT)+
pars[4]*CO2dif+pars[5]*log(MAT)*CO2dif
make_pct(y)
}
test$ESpct <- mod_fun(MAPi, test$MAT, CO2i, coef(modfit))
ggplot(test, aes(x = log(MAT), y = make_pct(es))) +
geom_line(aes(y=ESpct), color = "red", size=0.8) +
geom_point() + theme_classic()

Resources