I'm conducting data exploration on lots of parameters (including multiple ways to summarize a parameter). I'm trying to prioritize variables for prediction onto test data. I'd like to use dredge to examine all possible combinations (and two-way interactions) to narrow down the variables to explore in more depth. What's complicated is that I have several variables in different categories: diet, sea surface temperature (SST), and NAO and only want to include one variable from each category in a model at a time. Ultimately I want to see if variance of SST is more useful than average etc, so I don't want them in the same model (and they would likely be highly correlated as well). Currently, I've created global models for each combination of variables and dredged the global model. I want to view the outputs of the top model and all models within 2 AICc, across each of the global models I assessed. At first, I'd like to just view a list of the model formula and AIC of each of those top models. Then I'd like to be able to access the summaries of the top models for further exploration.
So my questions are how to save the desired outputs and if my approach is the best way to build the models and dredge them. Thank you!
My data named Model5_Diet_EnvData_SummarizedByRegionHY_RecruitmentSummaries_SubAbbrev (first 5 rows):
structure(list(DP = c(1, 1, 0, 0, 0), HI = structure(c(2L, 2L,
2L, 2L, 2L), levels = c("MR", "MSI", "SINWR"), class = "factor"),
NIP = c(1, 1, 1, 0, 0), PP = structure(c(2L, 1L, 1L, 1L,
1L), levels = c("0", "1"), class = "factor"), MPFHMOAP = c(0.750452915231012,
0.453965698407607, 0.441925608384321, 0.722610207052164,
0.441925608384321), MHVHMOAP = c(-0.0524813927346641, 1.01282281910627,
0.757232590524714, 0.99109247578557, 0.757232590524714),
MLVHMOAP = c(0.73056230124052, 1.59501115848116, 0.598198676596868,
0.449493863725604, 0.598198676596868), MPFHMORAP = c(0.523489867811386,
0.192245391224712, 0.178793838876015, 0.492383153005024,
0.178793838876015), MHVHMORAP = c(-0.329148414641852, 0.788691843441836,
0.520497046213404, 0.765889856632941, 0.520497046213404),
MLVHMORAP = c(0.688793519232913, 1.70845779144675, 0.532663451086749,
0.357258065885503, 0.532663451086749), MPFAAP = c(0.43070772291489,
1.13584913195813, -0.426234961686277, -0.203429863689975,
-0.426234961686277), MHVAAP = c(-0.0579325416076462, 0.849585831032181,
0.121001819024069, -0.201997642218822, 0.121001819024069),
MLVAAP = c(-1.35485853425042, 0.484252451276881, 0.768258290849834,
1.48519986147912, 0.768258290849834), MPFVAP = c(-0.244597407127982,
1.27217782014326, 0.247421666481899, 0.774014226262795, 0.247421666481899
), MHVVAP = c(-0.37754544487536, 0.758649910763721, 0.681804652562789,
0.887267156622061, 0.681804652562789), MLVVAP = c(-0.572980090345317,
1.62888586457038, -0.0115436727418632, 0.0160827313504334,
-0.0115436727418632), SSTANAP = c(1.07683379079566, -1.38798265446896,
-0.559237858951006, 0.0697567434994638, -0.559237858951006
), SSTVNAP = c(2.09070187356431, 0.144862160369841, -0.853503337708263,
-1.68867970657618, -0.853503337708263), SSTABAAP = c(0.173411254232863,
-1.11203442428642, -0.473195106714148, 0.827784658554605,
-0.473195106714148), SSTVBAAP = c(-0.629121122035084, 0.392528189012985,
0.379784122762904, 0.490232696930265, 0.379784122762904),
NAOA = c(0.897128780719261, -0.204661067596483, -1.061094353052,
-0.824996528412911, -1.061094353052), NAOV = c(-0.028733326081941,
0.855031609762443, -0.028481970866968, 0.766860571745026,
-0.028481970866968)), row.names = c(NA, -5L), class = c("tbl_df",
"tbl", "data.frame"))
Subset of the code I ran (my full data has ~70 global models):
# Build global models for all combinations of diet/sst/nao
model5_diet_HI_MPFHMOAP_SSTANAP_NAOA_DRsatr_autoglm=glm(DP~(HI+MPFHMOAP+SSTANAP+NAOA)^2, data = Model5_Diet_EnvData_SummarizedByRegionHY_RecruitmentSummaries_SubAbbrev, family = binomial(link = "logit"))
model5_diet_HI_MPFHMOAP_SSTVNAP_NAOA_DRsatr_autoglm=glm(DP~(HI+MPFHMOAP+SSTVNAP+NAOA)^2, data = Model5_Diet_EnvData_SummarizedByRegionHY_RecruitmentSummaries_SubAbbrev, family = binomial(link = "logit"))
model5_diet_HI_MPFHMOAP_SSTABAAP_NAOA_DRsatr_autoglm=glm(DP~(HI+MPFHMOAP+SSTABAAP+NAOA)^2, data = Model5_Diet_EnvData_SummarizedByRegionHY_RecruitmentSummaries_SubAbbrev, family = binomial(link = "logit"))
# Dredge global models
options(na.action = "na.fail") # Required for dredge to run
model5_diet_HI_MPFHMOAP_SSTANAP_NAOA_DRdredge=dredge(model5_diet_HI_MPFHMOAP_SSTANAP_NAOA_DRsatr_autoglm, beta = "none", evaluate = TRUE, rank = AICc)
model5_diet_HI_MPFHMOAP_SSTVNAP_NAOA_DRdredge=dredge(model5_diet_HI_MPFHMOAP_SSTVNAP_NAOA_DRsatr_autoglm, beta = "none", evaluate = TRUE, rank = AICc)
model5_diet_HI_MPFHMOAP_SSTABAAP_NAOA_DRdredge=dredge(model5_diet_HI_MPFHMOAP_SSTABAAP_NAOA_DRsatr_autoglm, beta = "none", evaluate = TRUE, rank = AICc)
options(na.action = "na.omit") # set back to default
Desired output: | Model | AIC | | -------- | -------------- | | DP ~ HI + NAOA + SSTANAP + HI:NAOA + HI:SSTANAP + NAOA:SSTANAP + 1 | 809.3 | | DP ~ HI + MPFHMOAP + NAOA + HI:MPFHMOAP + HI:NAOA + 1 | 810.6 | | DP ~ HI + MPFHMOAP + NAOA + SSTABAAP + HI:MPFHMOAP + HI:NAOA + HI:SSTABAAP + 1 | 810.8 |
Here is an approach to get your desired output consisting of models specified as character and accompanying AICc values.
Your theoretical questions (i.e., is this a good approach) are a better fit for Cross Validated.
options(na.action = "na.fail")
library(dplyr)
library(MuMIn)
library(purrr)
library(tidyr)
# a list containing the global models
globalmodels_topAIC <- list(
global_model_1 = glm(mpg ~ (cyl + disp + hp + drat)^2, data = mtcars),
global_model_2 = glm(mpg ~ (wt + qsec + vs + am)^2, data = mtcars)
) %>%
# for each global model dredge and make a tibble consisting of model and AICc
map(\(global_model) {
global_model %>%
dredge() %>%
as_tibble() %>%
# keep only predictor columns and AICc by deselecting all others
select(-c("(Intercept)", df, logLik, delta, weight)) %>%
rowwise() %>%
mutate(
# for predictor columns replace value with name of column if not missing
across(-AICc, ~ if_else(!is.na(.), cur_column(), NA_character_)),
intercept = 1
) %>%
# collapse values in predictor columns by removing NAs and separate with +
unite(model, -AICc, sep = " + ", remove = TRUE, na.rm = TRUE)
}) %>%
# row bind resulting list of tibbles
list_rbind() %>%
# sort by AICc
arrange(AICc)%>%
as.data.frame()
#> Fixed term is "(Intercept)"
#> Fixed term is "(Intercept)"
globalmodels_topAIC
#> model
#> 1 am + qsec + wt + am:wt + 1
#> 2 am + qsec + wt + am:wt + qsec:wt + 1
#> 3 am + qsec + wt + am:qsec + am:wt + 1
#> 4 am + qsec + vs + wt + am:wt + 1
#> 5 am + qsec + wt + am:qsec + am:wt + qsec:wt + 1
#> 6 am + qsec + vs + wt + am:wt + qsec:wt + 1
#> 7 am + qsec + vs + wt + am:wt + qsec:vs + 1
#> 8 am + qsec + wt + am:qsec + 1
#> 9 am + qsec + vs + wt + am:qsec + am:wt + 1
#> 10 am + qsec + vs + wt + am:vs + am:wt + 1
#> 11 am + qsec + vs + wt + am:vs + am:wt + qsec:wt + 1
#> 12 am + qsec + vs + wt + am:wt + vs:wt + 1
#> 13 am + vs + wt + am:wt + 1
#> 14 am + qsec + vs + wt + am:wt + qsec:wt + vs:wt + 1
#> 15 qsec + vs + wt + vs:wt + 1
#> 16 am + qsec + vs + wt + am:qsec + am:wt + qsec:wt + 1
#> 17 am + qsec + wt + 1
#> 18 am + qsec + vs + wt + am:wt + qsec:vs + qsec:wt + 1
#> 19 am + qsec + vs + wt + vs:wt + 1
#> 20 am + qsec + vs + wt + am:qsec + am:vs + am:wt + 1
#> 21 am + qsec + wt + am:qsec + qsec:wt + 1
#> 22 am + qsec + vs + wt + am:qsec + 1
#> 23 am + qsec + vs + wt + am:qsec + am:wt + qsec:vs + 1
#> 24 am + qsec + vs + wt + am:vs + am:wt + qsec:vs + 1
#> 25 am + qsec + wt + qsec:wt + 1
#> 26 am + qsec + vs + wt + am:wt + qsec:vs + vs:wt + 1
#> 27 qsec + wt + 1
#> 28 am + vs + wt + am:vs + am:wt + 1
#> 29 am + qsec + vs + wt + am:vs + am:wt + vs:wt + 1
#> 30 am + qsec + vs + wt + am:qsec + am:wt + vs:wt + 1
#> 31 am + vs + wt + am:wt + vs:wt + 1
#> 32 am + qsec + vs + wt + am:vs + am:wt + qsec:wt + vs:wt + 1
#> 33 am + qsec + vs + wt + am:vs + am:wt + qsec:vs + qsec:wt + 1
#> 34 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:wt + 1
#> 35 am + qsec + vs + wt + am:qsec + am:wt + qsec:wt + vs:wt + 1
#> 36 am + qsec + vs + wt + am:qsec + vs:wt + 1
#> 37 am + qsec + vs + wt + am:vs + 1
#> 38 qsec + vs + wt + qsec:vs + vs:wt + 1
#> 39 qsec + vs + wt + qsec:wt + vs:wt + 1
#> 40 vs + wt + vs:wt + 1
#> 41 am + qsec + vs + wt + 1
#> 42 qsec + wt + qsec:wt + 1
#> 43 am + wt + am:wt + 1
#> 44 am + qsec + vs + wt + am:wt + qsec:vs + qsec:wt + vs:wt + 1
#> 45 am + qsec + vs + wt + qsec:vs + vs:wt + 1
#> 46 am + qsec + vs + wt + qsec:vs + 1
#> 47 am + qsec + vs + wt + qsec:wt + vs:wt + 1
#> 48 am + qsec + vs + wt + am:vs + vs:wt + 1
#> 49 am + qsec + vs + wt + am:qsec + am:wt + qsec:vs + qsec:wt + 1
#> 50 am + qsec + vs + wt + am:qsec + am:vs + vs:wt + 1
#> 51 am + qsec + vs + wt + am:qsec + am:vs + am:wt + vs:wt + 1
#> 52 am + qsec + vs + wt + qsec:vs + qsec:wt + 1
#> 53 am + qsec + vs + wt + am:qsec + qsec:vs + 1
#> 54 qsec + vs + wt + 1
#> 55 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:vs + 1
#> 56 cyl + disp + hp + cyl:disp + 1
#> 57 am + qsec + vs + wt + am:qsec + am:vs + 1
#> 58 am + qsec + vs + wt + am:qsec + qsec:wt + 1
#> 59 am + qsec + vs + wt + qsec:wt + 1
#> 60 cyl + disp + cyl:disp + 1
#> 61 am + vs + wt + am:vs + am:wt + vs:wt + 1
#> 62 qsec + vs + wt + qsec:vs + 1
#> 63 am + qsec + vs + wt + am:qsec + am:wt + qsec:vs + vs:wt + 1
#> 64 am + qsec + vs + wt + am:vs + qsec:vs + 1
#> 65 disp + hp + disp:hp + 1
#> 66 am + qsec + vs + wt + am:vs + am:wt + qsec:vs + vs:wt + 1
#> 67 am + vs + wt + vs:wt + 1
#> 68 am + qsec + vs + wt + am:vs + qsec:wt + 1
#> 69 qsec + vs + wt + qsec:vs + qsec:wt + 1
#> 70 cyl + disp + hp + cyl:disp + cyl:hp + 1
#> 71 vs + wt + 1
#> 72 qsec + vs + wt + qsec:wt + 1
#> 73 am + qsec + vs + wt + am:qsec + am:wt + qsec:vs + qsec:wt + vs:wt + 1
#> 74 qsec + vs + wt + qsec:vs + qsec:wt + vs:wt + 1
#> 75 am + qsec + vs + wt + am:qsec + qsec:wt + vs:wt + 1
#> 76 am + qsec + vs + wt + am:qsec + qsec:vs + vs:wt + 1
#> 77 am + qsec + vs + wt + am:vs + am:wt + qsec:vs + qsec:wt + vs:wt + 1
#> 78 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:wt + vs:wt + 1
#> 79 am + qsec + vs + wt + qsec:vs + qsec:wt + vs:wt + 1
#> 80 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:vs + qsec:wt + 1
#> 81 am + qsec + vs + wt + am:vs + qsec:vs + vs:wt + 1
#> 82 cyl + disp + hp + cyl:disp + disp:hp + 1
#> 83 am + qsec + vs + wt + am:vs + qsec:wt + vs:wt + 1
#> 84 am + vs + wt + 1
#> 85 cyl + disp + drat + hp + cyl:disp + 1
#> 86 am + qsec + vs + wt + am:qsec + am:vs + qsec:wt + vs:wt + 1
#> 87 am + vs + wt + am:vs + 1
#> 88 am + qsec + vs + wt + am:qsec + qsec:vs + qsec:wt + 1
#> 89 cyl + disp + drat + cyl:disp + 1
#> 90 cyl + disp + hp + cyl:hp + 1
#> 91 am + qsec + vs + wt + am:vs + qsec:vs + qsec:wt + 1
#> 92 am + qsec + vs + wt + am:qsec + am:vs + qsec:vs + vs:wt + 1
#> 93 am + qsec + vs + wt + am:qsec + am:vs + qsec:vs + 1
#> 94 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:vs + vs:wt + 1
#> 95 cyl + disp + hp + disp:hp + 1
#> 96 am + qsec + vs + wt + am:qsec + am:vs + qsec:wt + 1
#> 97 disp + drat + hp + disp:hp + 1
#> 98 am + vs + wt + am:vs + vs:wt + 1
#> 99 cyl + disp + hp + cyl:disp + cyl:hp + disp:hp + 1
#> 100 cyl + disp + drat + hp + cyl:disp + cyl:hp + 1
#> 101 wt + 1
#> 102 am + qsec + vs + wt + am:qsec + qsec:vs + qsec:wt + vs:wt + 1
#> 103 cyl + disp + hp + cyl:hp + disp:hp + 1
#> 104 am + qsec + vs + wt + am:vs + qsec:vs + qsec:wt + vs:wt + 1
#> 105 cyl + disp + drat + cyl:disp + disp:drat + 1
#> 106 cyl + disp + drat + hp + cyl:disp + cyl:drat + 1
#> 107 cyl + disp + drat + hp + cyl:disp + drat:hp + 1
#> 108 cyl + disp + drat + cyl:disp + cyl:drat + 1
#> 109 cyl + disp + drat + hp + cyl:disp + disp:hp + 1
#> 110 cyl + disp + drat + hp + cyl:disp + disp:drat + 1
#> 111 cyl + disp + drat + hp + cyl:hp + 1
#> 112 am + qsec + vs + wt + am:qsec + am:vs + am:wt + qsec:vs + qsec:wt + vs:wt + 1
#> 113 disp + drat + hp + disp:drat + disp:hp + 1
#> 114 cyl + disp + drat + hp + disp:hp + 1
#> 115 disp + drat + disp:drat + 1
#> 116 disp + drat + hp + disp:hp + drat:hp + 1
#> 117 cyl + hp + cyl:hp + 1
#> 118 am + qsec + vs + wt + am:qsec + am:vs + qsec:vs + qsec:wt + 1
#> 119 cyl + disp + 1
#> 120 am + qsec + vs + wt + am:qsec + am:vs + qsec:vs + qsec:wt + vs:wt + 1
#> 121 disp + drat + hp + 1
#> 122 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + 1
#> 123 cyl + disp + drat + hp + cyl:disp + cyl:hp + drat:hp + 1
#> 124 am + wt + 1
#> 125 cyl + drat + hp + cyl:hp + 1
#> 126 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:hp + 1
#> 127 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:drat + 1
#> 128 cyl + disp + drat + disp:drat + 1
#> 129 cyl + disp + drat + cyl:drat + 1
#> 130 disp + drat + hp + disp:drat + 1
#> 131 disp + hp + 1
#> 132 cyl + drat + hp + 1
#> 133 cyl + 1
#> 134 cyl + disp + hp + 1
#> 135 cyl + disp + drat + hp + cyl:hp + disp:drat + 1
#> 136 cyl + disp + drat + 1
#> 137 cyl + disp + drat + hp + cyl:hp + disp:hp + 1
#> 138 disp + drat + hp + drat:hp + 1
#> 139 cyl + disp + drat + cyl:disp + cyl:drat + disp:drat + 1
#> 140 cyl + disp + drat + hp + cyl:drat + cyl:hp + 1
#> 141 drat + hp + 1
#> 142 cyl + disp + drat + hp + cyl:disp + disp:hp + drat:hp + 1
#> 143 cyl + hp + 1
#> 144 disp + 1
#> 145 cyl + drat + 1
#> 146 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:hp + 1
#> 147 cyl + disp + drat + hp + cyl:disp + disp:drat + disp:hp + 1
#> 148 cyl + disp + drat + hp + cyl:hp + drat:hp + 1
#> 149 cyl + disp + drat + hp + 1
#> 150 cyl + drat + cyl:drat + 1
#> 151 cyl + disp + drat + hp + cyl:disp + cyl:drat + drat:hp + 1
#> 152 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:drat + 1
#> 153 cyl + disp + drat + hp + cyl:drat + disp:hp + 1
#> 154 cyl + disp + drat + hp + cyl:disp + disp:drat + drat:hp + 1
#> 155 disp + drat + hp + disp:drat + disp:hp + drat:hp + 1
#> 156 cyl + disp + drat + hp + disp:drat + disp:hp + 1
#> 157 cyl + disp + drat + hp + disp:hp + drat:hp + 1
#> 158 cyl + drat + hp + cyl:drat + 1
#> 159 disp + drat + 1
#> 160 cyl + disp + drat + hp + cyl:drat + 1
#> 161 cyl + drat + hp + drat:hp + 1
#> 162 cyl + disp + drat + hp + disp:drat + 1
#> 163 cyl + drat + hp + cyl:drat + cyl:hp + 1
#> 164 cyl + drat + hp + cyl:hp + drat:hp + 1
#> 165 cyl + disp + drat + cyl:drat + disp:drat + 1
#> 166 cyl + disp + drat + hp + drat:hp + 1
#> 167 disp + drat + hp + disp:drat + drat:hp + 1
#> 168 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:hp + 1
#> 169 cyl + disp + drat + hp + cyl:hp + disp:drat + drat:hp + 1
#> 170 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:drat + drat:hp + 1
#> 171 drat + hp + drat:hp + 1
#> 172 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:hp + drat:hp + 1
#> 173 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:drat + 1
#> 174 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + drat:hp + 1
#> 175 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:drat + disp:hp + 1
#> 176 cyl + disp + drat + hp + cyl:drat + disp:hp + drat:hp + 1
#> 177 cyl + disp + drat + hp + cyl:hp + disp:drat + disp:hp + 1
#> 178 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:hp + 1
#> 179 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:drat + 1
#> 180 cyl + disp + drat + hp + cyl:hp + disp:hp + drat:hp + 1
#> 181 cyl + disp + drat + hp + cyl:drat + cyl:hp + drat:hp + 1
#> 182 cyl + disp + drat + hp + disp:drat + disp:hp + drat:hp + 1
#> 183 cyl + disp + drat + hp + cyl:drat + disp:drat + disp:hp + 1
#> 184 cyl + disp + drat + hp + cyl:disp + disp:drat + disp:hp + drat:hp + 1
#> 185 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:hp + drat:hp + 1
#> 186 cyl + drat + hp + cyl:drat + drat:hp + 1
#> 187 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:drat + disp:hp + 1
#> 188 am + qsec + vs + am:qsec + 1
#> 189 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:drat + drat:hp + 1
#> 190 cyl + disp + drat + hp + cyl:drat + disp:drat + 1
#> 191 cyl + disp + drat + hp + cyl:drat + drat:hp + 1
#> 192 cyl + disp + drat + hp + disp:drat + drat:hp + 1
#> 193 am + qsec + am:qsec + 1
#> 194 cyl + drat + hp + cyl:drat + cyl:hp + drat:hp + 1
#> 195 am + qsec + vs + 1
#> 196 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:drat + disp:hp + 1
#> 197 am + qsec + 1
#> 198 am + vs + 1
#> 199 cyl + disp + drat + hp + cyl:hp + disp:drat + disp:hp + drat:hp + 1
#> 200 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:drat + drat:hp + 1
#> 201 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:hp + drat:hp + 1
#> 202 am + qsec + vs + am:vs + 1
#> 203 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:drat + drat:hp + 1
#> 204 cyl + disp + drat + hp + cyl:disp + cyl:hp + disp:drat + disp:hp + drat:hp + 1
#> 205 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:hp + drat:hp + 1
#> 206 cyl + disp + drat + hp + cyl:drat + disp:drat + disp:hp + drat:hp + 1
#> 207 am + qsec + vs + am:qsec + am:vs + 1
#> 208 am + vs + am:vs + 1
#> 209 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:drat + disp:hp + 1
#> 210 am + qsec + vs + qsec:vs + 1
#> 211 am + qsec + vs + am:qsec + qsec:vs + 1
#> 212 cyl + disp + drat + hp + cyl:drat + disp:drat + drat:hp + 1
#> 213 cyl + disp + drat + hp + cyl:disp + cyl:drat + disp:drat + disp:hp + drat:hp + 1
#> 214 am + qsec + vs + am:vs + qsec:vs + 1
#> 215 cyl + disp + drat + hp + cyl:disp + cyl:drat + cyl:hp + disp:drat + disp:hp + drat:hp + 1
#> 216 cyl + disp + drat + hp + cyl:drat + cyl:hp + disp:drat + disp:hp + drat:hp + 1
#> 217 am + qsec + vs + am:qsec + am:vs + qsec:vs + 1
#> 218 hp + 1
#> 219 drat + 1
#> 220 vs + 1
#> 221 qsec + vs + 1
#> 222 am + 1
#> 223 qsec + vs + qsec:vs + 1
#> 224 qsec + 1
#> 225 1
#> 226 1
#> AICc
#> 1 147.7336
#> 2 149.0083
#> 3 150.8206
#> 4 151.0402
#> 5 152.2860
#> 6 152.5927
#> 7 154.1952
#> 8 154.3050
#> 9 154.4147
#> 10 154.5234
#> 11 154.5632
#> 12 154.6156
#> 13 155.3045
#> 14 155.5000
#> 15 155.9726
#> 16 156.1921
#> 17 156.4271
#> 18 156.4529
#> 19 156.7605
#> 20 157.3624
#> 21 157.6063
#> 22 157.6106
#> 23 157.6837
#> 24 158.0734
#> 25 158.0844
#> 26 158.1030
#> 27 158.2020
#> 28 158.2414
#> 29 158.2583
#> 30 158.3340
#> 31 158.5707
#> 32 158.6087
#> 33 158.6446
#> 34 158.7698
#> 35 158.8777
#> 36 158.9134
#> 37 159.0775
#> 38 159.1334
#> 39 159.2779
#> 40 159.4457
#> 41 159.4793
#> 42 159.5843
#> 43 159.7837
#> 44 159.7933
#> 45 159.8672
#> 46 160.0380
#> 47 160.3219
#> 48 160.3545
#> 49 160.4727
#> 50 160.5169
#> 51 160.6676
#> 52 160.7307
#> 53 160.8759
#> 54 161.0159
#> 55 161.0339
#> 56 161.0998
#> 57 161.1938
#> 58 161.1991
#> 59 161.3821
#> 60 161.4571
#> 61 161.7869
#> 62 161.9377
#> 63 161.9770
#> 64 162.0851
#> 65 162.2164
#> 66 162.2678
#> 67 162.3081
#> 68 162.4574
#> 69 162.5145
#> 70 162.5532
#> 71 162.5761
#> 72 162.6361
#> 73 162.6531
#> 74 162.6545
#> 75 162.7628
#> 76 162.7683
#> 77 163.1019
#> 78 163.3192
#> 79 163.3466
#> 80 163.3592
#> 81 163.7302
#> 82 163.9840
#> 83 164.2428
#> 84 164.2719
#> 85 164.2808
#> 86 164.2855
#> 87 164.2942
#> 88 164.3435
#> 89 164.3818
#> 90 164.6133
#> 91 164.6221
#> 92 164.6745
#> 93 164.7967
#> 94 164.9646
#> 95 165.0930
#> 96 165.1087
#> 97 165.2344
#> 98 165.6142
#> 99 165.9542
#> 100 166.1438
#> 101 166.8866
#> 102 167.0515
#> 103 167.1616
#> 104 167.3447
#> 105 167.3986
#> 106 167.4436
#> 107 167.5470
#> 108 167.5763
#> 109 167.5770
#> 110 167.5883
#> 111 167.7054
#> 112 167.8586
#> 113 168.3047
#> 114 168.3545
#> 115 168.5005
#> 116 168.5043
#> 117 168.5943
#> 118 168.6188
#> 119 168.6271
#> 120 168.8845
#> 121 169.3284
#> 122 169.4279
#> 123 169.4288
#> 124 169.5107
#> 125 169.5468
#> 126 169.7700
#> 127 169.8009
#> 128 169.8376
#> 129 169.9034
#> 130 169.9637
#> 131 170.1000
#> 132 170.1171
#> 133 170.1636
#> 134 170.3261
#> 135 170.6409
#> 136 170.7090
#> 137 170.7554
#> 138 170.8182
#> 139 170.9187
#> 140 170.9372
#> 141 170.9861
#> 142 171.0134
#> 143 171.0433
#> 144 171.0665
#> 145 171.1001
#> 146 171.1473
#> 147 171.1649
#> 148 171.1978
#> 149 171.2960
#> 150 171.3503
#> 151 171.3601
#> 152 171.3644
#> 153 171.4444
#> 154 171.4650
#> 155 171.5689
#> 156 171.7663
#> 157 171.9437
#> 158 172.0425
#> 159 172.2185
#> 160 172.3616
#> 161 172.4311
#> 162 172.4918
#> 163 172.7735
#> 164 172.8513
#> 165 172.9799
#> 166 173.0097
#> 167 173.0658
#> 168 173.0757
#> 169 173.2757
#> 170 173.4749
#> 171 173.4991
#> 172 173.5216
#> 173 173.6485
#> 174 173.6562
#> 175 173.8606
#> 176 174.2266
#> 177 174.3431
#> 178 174.3893
#> 179 174.5019
#> 180 174.6701
#> 181 174.7195
#> 182 175.1133
#> 183 175.2016
#> 184 175.2960
#> 185 175.3076
#> 186 175.3230
#> 187 175.4209
#> 188 175.5767
#> 189 175.6484
#> 190 175.8848
#> 191 175.9479
#> 192 176.0391
#> 193 176.0943
#> 194 176.2263
#> 195 176.3883
#> 196 176.5857
#> 197 177.0837
#> 198 177.1610
#> 199 177.3924
#> 200 177.5698
#> 201 177.6959
#> 202 177.8747
#> 203 177.9548
#> 204 177.9863
#> 205 178.0329
#> 206 178.4071
#> 207 178.4757
#> 208 178.5041
#> 209 178.6373
#> 210 178.6508
#> 211 178.8583
#> 212 179.6764
#> 213 180.0194
#> 214 181.1160
#> 215 181.7312
#> 216 182.0494
#> 217 182.0697
#> 218 182.0958
#> 219 191.6571
#> 220 193.0043
#> 221 194.8837
#> 222 197.3415
#> 223 197.6783
#> 224 205.4453
#> 225 209.1693
#> 226 209.1693
Created on 2024-09-05 with reprex v2.1.1