simple_regressions.R

###load libraries and read truncated indices data####
suppressPackageStartupMessages(library(tidyverse))

## Warning: package 'tidyverse' was built under R version 3.3.3

## Warning: package 'purrr' was built under R version 3.3.3

library(broom)
library(readr)
library(knitr)

setwd ("P:/MTM2 Oranga Taiao/EMP Data/EMP Data/Biotic Index calculator/Index calculation/MTM new Data")
source('theme_javier.R')
theme_set(theme_javier(8))
options(digits = 3)

all_ind<- read_csv('all_indices_truncated.csv',col_types =  cols(Date = 'D')) %>% 
  mutate(Type = factor(Type),
         logTOC = log(TOC+1))
dat_simple <- 
  all_ind %>%
  gather(index, value,AMBI:MEDOCC,BQI:TBI,logN,S) 

###Mud simple regression
simple_mud <- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index2 = lm(value ~ sqrtmud,data = .,weights = sqrt(n)))

coef_mud <- tidy(simple_mud, fit_index2,conf.int = T)
summ_mud <- glance(simple_mud, fit_index2)
kable(coef_mud)

index	term	estimate	std.error	statistic	conf.low	conf.high
AMBI	(Intercept)	1.378	0.051	27.25	1.279	1.478
AMBI	sqrtmud	0.170	0.012	14.61	0.147	0.193
AMBI_S	(Intercept)	1.446	0.036	39.75	1.374	1.517
AMBI_S	sqrtmud	0.162	0.008	19.45	0.146	0.179
BENTIX	(Intercept)	5.204	0.078	66.63	5.051	5.358
BENTIX	sqrtmud	-0.232	0.018	-12.93	-0.267	-0.197
BQI	(Intercept)	6.394	0.112	57.19	6.175	6.614
BQI	sqrtmud	-0.335	0.026	-13.03	-0.385	-0.284
ITI	(Intercept)	34.826	1.122	31.03	32.622	37.031
ITI	sqrtmud	-1.360	0.258	-5.28	-1.866	-0.853
logN	(Intercept)	4.613	0.067	68.48	4.480	4.745
logN	sqrtmud	-0.085	0.015	-5.49	-0.115	-0.055
M_AMBI	(Intercept)	0.646	0.008	81.34	0.630	0.662
M_AMBI	sqrtmud	-0.027	0.002	-14.91	-0.031	-0.024
MEDOCC	(Intercept)	1.857	0.066	28.25	1.728	1.986
MEDOCC	sqrtmud	0.209	0.015	13.87	0.180	0.239
S	(Intercept)	14.590	0.327	44.62	13.948	15.232
S	sqrtmud	-0.865	0.075	-11.52	-1.012	-0.717
TBI	(Intercept)	0.391	0.010	40.77	0.372	0.410
TBI	sqrtmud	-0.023	0.002	-10.40	-0.027	-0.019

kable(summ_mud)

index	r.squared	adj.r.squared	sigma	statistic	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.284	0.283	1.013	213.3	2	-504	1014	1027	551.6	537
AMBI_S	0.413	0.412	0.729	378.3	2	-326	658	671	285.0	537
BENTIX	0.237	0.236	1.565	167.2	2	-738	1482	1495	1314.9	537
BQI	0.240	0.239	2.240	169.8	2	-931	1869	1882	2694.2	537
ITI	0.049	0.047	22.484	27.8	2	-2174	4355	4368	271458.9	537
logN	0.053	0.051	1.349	30.1	2	-658	1323	1335	977.9	537
M_AMBI	0.293	0.292	0.159	222.4	2	494	-982	-969	13.6	537
MEDOCC	0.264	0.262	1.316	192.4	2	-645	1296	1309	930.6	537
S	0.198	0.197	6.551	132.6	2	-1510	3026	3038	23042.7	537
TBI	0.168	0.166	0.192	108.1	2	392	-778	-765	19.8	537

###Metals simple regression
simple_metals <- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index1 = lm(value ~ metals,data = .,weights = sqrt(n)))

coef_metals <- tidy(simple_metals, fit_index1,conf.int = T)
summ_metals <- glance(simple_metals, fit_index1)
kable(coef_metals)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.852	0.038	48.65	0.000	1.777	1.927
AMBI	metals	0.063	0.014	4.44	0.000	0.035	0.090
AMBI_S	(Intercept)	1.929	0.028	68.68	0.000	1.874	1.985
AMBI_S	metals	0.065	0.010	6.30	0.000	0.045	0.086
BENTIX	(Intercept)	4.598	0.058	78.76	0.000	4.483	4.713
BENTIX	metals	-0.073	0.022	-3.38	0.001	-0.116	-0.030
BQI	(Intercept)	4.869	0.089	54.90	0.000	4.695	5.044
BQI	metals	-0.162	0.033	-4.95	0.000	-0.227	-0.098
ITI	(Intercept)	31.610	0.788	40.13	0.000	30.059	33.160
ITI	metals	0.332	0.291	1.14	0.255	-0.241	0.906
logN	(Intercept)	4.231	0.056	75.96	0.000	4.122	4.341
logN	metals	-0.062	0.021	-3.00	0.003	-0.102	-0.021
M_AMBI	(Intercept)	0.538	0.006	91.65	0.000	0.526	0.549
M_AMBI	metals	-0.013	0.002	-6.05	0.000	-0.017	-0.009
MEDOCC	(Intercept)	2.426	0.049	49.36	0.000	2.329	2.523
MEDOCC	metals	0.068	0.018	3.73	0.000	0.032	0.104
S	(Intercept)	10.745	0.249	43.16	0.000	10.255	11.235
S	metals	-0.465	0.092	-5.04	0.000	-0.646	-0.283
TBI	(Intercept)	0.278	0.007	42.05	0.000	0.265	0.292
TBI	metals	-0.011	0.002	-4.68	0.000	-0.016	-0.007

kable(coef_metals)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.852	0.038	48.65	0.000	1.777	1.927
AMBI	metals	0.063	0.014	4.44	0.000	0.035	0.090
AMBI_S	(Intercept)	1.929	0.028	68.68	0.000	1.874	1.985
AMBI_S	metals	0.065	0.010	6.30	0.000	0.045	0.086
BENTIX	(Intercept)	4.598	0.058	78.76	0.000	4.483	4.713
BENTIX	metals	-0.073	0.022	-3.38	0.001	-0.116	-0.030
BQI	(Intercept)	4.869	0.089	54.90	0.000	4.695	5.044
BQI	metals	-0.162	0.033	-4.95	0.000	-0.227	-0.098
ITI	(Intercept)	31.610	0.788	40.13	0.000	30.059	33.160
ITI	metals	0.332	0.291	1.14	0.255	-0.241	0.906
logN	(Intercept)	4.231	0.056	75.96	0.000	4.122	4.341
logN	metals	-0.062	0.021	-3.00	0.003	-0.102	-0.021
M_AMBI	(Intercept)	0.538	0.006	91.65	0.000	0.526	0.549
M_AMBI	metals	-0.013	0.002	-6.05	0.000	-0.017	-0.009
MEDOCC	(Intercept)	2.426	0.049	49.36	0.000	2.329	2.523
MEDOCC	metals	0.068	0.018	3.73	0.000	0.032	0.104
S	(Intercept)	10.745	0.249	43.16	0.000	10.255	11.235
S	metals	-0.465	0.092	-5.04	0.000	-0.646	-0.283
TBI	(Intercept)	0.278	0.007	42.05	0.000	0.265	0.292
TBI	metals	-0.011	0.002	-4.68	0.000	-0.016	-0.007

###sqrtTP simple regression
simple_sqrtTP<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index3 = lm(value ~ sqrtTP,data = .,weights = sqrt(n)))

coef_sqrtTP <- tidy(simple_sqrtTP, fit_index3,conf.int = T)
summ_sqrtTP <-  glance(simple_sqrtTP, fit_index3)
kable(coef_sqrtTP)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.044	0.165	6.33	0.000	0.719	1.369
AMBI	sqrtTP	0.044	0.009	5.04	0.000	0.027	0.061
AMBI_S	(Intercept)	1.111	0.121	9.20	0.000	0.873	1.349
AMBI_S	sqrtTP	0.044	0.006	6.95	0.000	0.032	0.057
BENTIX	(Intercept)	5.372	0.258	20.81	0.000	4.864	5.881
BENTIX	sqrtTP	-0.040	0.014	-2.96	0.003	-0.067	-0.013
BQI	(Intercept)	6.345	0.415	15.28	0.000	5.527	7.164
BQI	sqrtTP	-0.083	0.022	-3.80	0.000	-0.126	-0.040
ITI	(Intercept)	25.438	3.384	7.52	0.000	18.772	32.104
ITI	sqrtTP	0.394	0.177	2.22	0.027	0.045	0.744
logN	(Intercept)	4.663	0.269	17.36	0.000	4.134	5.192
logN	sqrtTP	-0.025	0.014	-1.81	0.071	-0.053	0.002
M_AMBI	(Intercept)	0.670	0.027	24.84	0.000	0.616	0.723
M_AMBI	sqrtTP	-0.007	0.001	-5.07	0.000	-0.010	-0.004
MEDOCC	(Intercept)	1.482	0.215	6.88	0.000	1.058	1.906
MEDOCC	sqrtTP	0.051	0.011	4.52	0.000	0.029	0.073
S	(Intercept)	14.908	1.172	12.72	0.000	12.599	17.217
S	sqrtTP	-0.230	0.061	-3.74	0.000	-0.351	-0.109
TBI	(Intercept)	0.373	0.030	12.32	0.000	0.313	0.432
TBI	sqrtTP	-0.006	0.002	-3.47	0.001	-0.009	-0.002

kable(summ_sqrtTP)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.096	0.092	0.972	25.44	0.000	2	-222	450	461	226.94	240
AMBI_S	0.168	0.164	0.712	48.30	0.000	2	-147	299	310	121.60	240
BENTIX	0.035	0.031	1.521	8.76	0.003	2	-330	667	677	555.20	240
BQI	0.057	0.053	2.448	14.41	0.000	2	-446	897	908	1437.92	240
ITI	0.020	0.016	19.940	4.94	0.027	2	-953	1912	1923	95426.55	240
logN	0.013	0.009	1.582	3.28	0.071	2	-340	686	697	600.94	240
M_AMBI	0.097	0.093	0.159	25.69	0.000	2	216	-427	-416	6.05	240
MEDOCC	0.079	0.075	1.269	20.45	0.000	2	-287	579	590	386.22	240
S	0.055	0.051	6.908	14.02	0.000	2	-697	1399	1410	11451.69	240
TBI	0.048	0.044	0.178	12.02	0.001	2	188	-371	-360	7.63	240

###Region simple regression
simple_region<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index4 = lm(value ~ Region,data = .,weights = sqrt(n)))

coef_region <- tidy(simple_region, fit_index4,conf.int = T)
summ_region <- glance(simple_region, fit_index4)
kable(coef_region)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.884	0.090	20.952	0.000	1.707	2.061
AMBI	RegionEastern North Island	-0.024	0.134	-0.178	0.858	-0.288	0.240
AMBI	RegionNorth Eastern	0.140	0.101	1.387	0.166	-0.058	0.338
AMBI	RegionSouthern	0.305	0.114	2.669	0.008	0.081	0.530
AMBI	RegionWestern North Island	0.061	0.120	0.510	0.610	-0.174	0.296
AMBI_S	(Intercept)	1.947	0.071	27.380	0.000	1.807	2.087
AMBI_S	RegionEastern North Island	0.048	0.106	0.447	0.655	-0.161	0.256
AMBI_S	RegionNorth Eastern	0.130	0.080	1.625	0.105	-0.027	0.286
AMBI_S	RegionSouthern	0.253	0.090	2.798	0.005	0.075	0.431
AMBI_S	RegionWestern North Island	-0.041	0.095	-0.437	0.663	-0.227	0.145
BENTIX	(Intercept)	4.547	0.133	34.310	0.000	4.287	4.807
BENTIX	RegionEastern North Island	0.405	0.198	2.044	0.041	0.016	0.794
BENTIX	RegionNorth Eastern	-0.226	0.149	-1.522	0.129	-0.518	0.066
BENTIX	RegionSouthern	-0.437	0.169	-2.596	0.010	-0.768	-0.106
BENTIX	RegionWestern North Island	-0.412	0.176	-2.335	0.020	-0.758	-0.065
BQI	(Intercept)	5.128	0.166	30.934	0.000	4.803	5.454
BQI	RegionEastern North Island	-1.788	0.248	-7.220	0.000	-2.275	-1.302
BQI	RegionNorth Eastern	0.427	0.186	2.298	0.022	0.062	0.792
BQI	RegionSouthern	-0.815	0.211	-3.868	0.000	-1.229	-0.401
BQI	RegionWestern North Island	0.975	0.221	4.422	0.000	0.542	1.408
ITI	(Intercept)	29.885	1.646	18.155	0.000	26.651	33.119
ITI	RegionEastern North Island	13.064	2.460	5.312	0.000	8.233	17.896
ITI	RegionNorth Eastern	-0.916	1.845	-0.496	0.620	-4.541	2.709
ITI	RegionSouthern	-0.303	2.093	-0.145	0.885	-4.414	3.809
ITI	RegionWestern North Island	-5.897	2.190	-2.693	0.007	-10.199	-1.596
logN	(Intercept)	3.993	0.094	42.648	0.000	3.809	4.177
logN	RegionEastern North Island	-0.755	0.140	-5.394	0.000	-1.029	-0.480
logN	RegionNorth Eastern	0.504	0.105	4.800	0.000	0.298	0.710
logN	RegionSouthern	0.514	0.119	4.317	0.000	0.280	0.748
logN	RegionWestern North Island	0.334	0.125	2.679	0.008	0.089	0.578
M_AMBI	(Intercept)	0.554	0.013	41.533	0.000	0.528	0.580
M_AMBI	RegionEastern North Island	-0.085	0.020	-4.287	0.000	-0.125	-0.046
M_AMBI	RegionNorth Eastern	0.011	0.015	0.731	0.465	-0.018	0.040
M_AMBI	RegionSouthern	-0.067	0.017	-3.940	0.000	-0.100	-0.034
M_AMBI	RegionWestern North Island	0.040	0.018	2.276	0.023	0.006	0.075
MEDOCC	(Intercept)	2.488	0.115	21.674	0.000	2.263	2.714
MEDOCC	RegionEastern North Island	-0.141	0.172	-0.824	0.411	-0.478	0.196
MEDOCC	RegionNorth Eastern	0.165	0.129	1.280	0.201	-0.088	0.417
MEDOCC	RegionSouthern	0.398	0.146	2.726	0.007	0.111	0.685
MEDOCC	RegionWestern North Island	0.089	0.153	0.585	0.559	-0.211	0.389
S	(Intercept)	11.085	0.496	22.366	0.000	10.111	12.059
S	RegionEastern North Island	-4.596	0.741	-6.206	0.000	-6.051	-3.141
S	RegionNorth Eastern	1.410	0.556	2.538	0.011	0.319	2.502
S	RegionSouthern	-1.340	0.630	-2.126	0.034	-2.578	-0.102
S	RegionWestern North Island	2.173	0.659	3.295	0.001	0.878	3.468
TBI	(Intercept)	0.302	0.014	22.167	0.000	0.275	0.328
TBI	RegionEastern North Island	-0.138	0.020	-6.806	0.000	-0.178	-0.098
TBI	RegionNorth Eastern	0.049	0.015	3.188	0.002	0.019	0.079
TBI	RegionSouthern	-0.072	0.017	-4.189	0.000	-0.106	-0.039
TBI	RegionWestern North Island	0.058	0.018	3.203	0.001	0.022	0.094

kable(summ_region)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.021	0.014	1.189	2.86	0.023	5	-588	1189	1214	754.5	534
AMBI_S	0.029	0.021	0.940	3.92	0.004	5	-462	936	962	472.0	534
BENTIX	0.050	0.042	1.752	6.95	0.000	5	-797	1607	1633	1639.0	534
BQI	0.277	0.271	2.192	51.09	0.000	5	-918	1848	1874	2564.6	534
ITI	0.114	0.108	21.760	17.25	0.000	5	-2155	4323	4348	252856.9	534
logN	0.208	0.202	1.238	35.07	0.000	5	-610	1232	1258	817.9	534
M_AMBI	0.136	0.130	0.176	21.02	0.000	5	440	-868	-842	16.6	534
MEDOCC	0.027	0.020	1.517	3.73	0.005	5	-720	1452	1478	1229.7	534
S	0.202	0.196	6.552	33.84	0.000	5	-1508	3029	3054	22922.3	534
TBI	0.275	0.269	0.180	50.62	0.000	5	429	-847	-821	17.3	534

###AFDW simple regression
simple_logAFDW<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index5 = lm(value ~ logAFDW,data = .,weights = sqrt(n)))

coef_logAFDW <- tidy(simple_logAFDW, fit_index5,conf.int = T)
summ_logAFDW <- glance(simple_logAFDW, fit_index5)
kable(coef_logAFDW)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.738	0.036	48.881	0.000	1.668	1.808
AMBI	logAFDW	0.342	0.044	7.798	0.000	0.256	0.429
AMBI_S	(Intercept)	1.776	0.026	67.280	0.000	1.724	1.828
AMBI_S	logAFDW	0.372	0.033	11.406	0.000	0.308	0.436
BENTIX	(Intercept)	4.669	0.058	80.191	0.000	4.554	4.783
BENTIX	logAFDW	-0.447	0.072	-6.214	0.000	-0.588	-0.305
BQI	(Intercept)	5.617	0.089	62.953	0.000	5.442	5.793
BQI	logAFDW	-0.468	0.110	-4.251	0.000	-0.685	-0.252
ITI	(Intercept)	31.128	0.858	36.301	0.000	29.442	32.814
ITI	logAFDW	-2.164	1.059	-2.043	0.042	-4.246	-0.082
logN	(Intercept)	4.286	0.048	89.957	0.000	4.193	4.380
logN	logAFDW	0.020	0.059	0.342	0.733	-0.096	0.136
M_AMBI	(Intercept)	0.586	0.006	97.208	0.000	0.574	0.598
M_AMBI	logAFDW	-0.045	0.007	-6.076	0.000	-0.060	-0.031
MEDOCC	(Intercept)	2.296	0.046	50.326	0.000	2.206	2.386
MEDOCC	logAFDW	0.414	0.056	7.347	0.000	0.303	0.525
S	(Intercept)	12.413	0.260	47.659	0.000	11.901	12.925
S	logAFDW	-0.949	0.322	-2.951	0.003	-1.582	-0.317
TBI	(Intercept)	0.342	0.008	44.489	0.000	0.327	0.357
TBI	logAFDW	-0.025	0.010	-2.682	0.008	-0.044	-0.007

kable(summ_logAFDW)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.133	0.131	0.948	60.805	0.000	2	-357	721	733	356.9	397
AMBI_S	0.247	0.245	0.704	130.093	0.000	2	-239	483	495	196.9	397
BENTIX	0.089	0.086	1.553	38.609	0.000	2	-554	1115	1126	957.2	397
BQI	0.044	0.041	2.380	18.069	0.000	2	-725	1455	1467	2248.5	397
ITI	0.010	0.008	22.871	4.175	0.042	2	-1627	3261	3273	207663.7	397
logN	0.000	-0.002	1.271	0.117	0.733	2	-474	955	967	641.2	397
M_AMBI	0.085	0.083	0.161	36.915	0.000	2	350	-695	-683	10.3	397
MEDOCC	0.120	0.117	1.217	53.982	0.000	2	-457	920	932	587.8	397
S	0.021	0.019	6.946	8.710	0.003	2	-1152	2310	2322	19156.4	397
TBI	0.018	0.015	0.205	7.196	0.008	2	253	-500	-488	16.7	397

###logPb simple regression
simple_logPb<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index6 = lm(value ~ logPb,data = .,weights = sqrt(n)))

coef_logPb <- tidy(simple_logPb, fit_index6,conf.int = T)
summ_logPb <- glance(simple_logPb, fit_index6)
kable(coef_logPb)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.449	0.098	14.71	0.000	1.255	1.643
AMBI	logPb	0.224	0.049	4.57	0.000	0.127	0.320
AMBI_S	(Intercept)	1.515	0.073	20.83	0.000	1.372	1.658
AMBI_S	logPb	0.231	0.036	6.37	0.000	0.159	0.302
BENTIX	(Intercept)	5.024	0.152	33.11	0.000	4.725	5.323
BENTIX	logPb	-0.238	0.076	-3.15	0.002	-0.386	-0.089
BQI	(Intercept)	5.883	0.230	25.58	0.000	5.430	6.336
BQI	logPb	-0.564	0.114	-4.93	0.000	-0.789	-0.339
ITI	(Intercept)	28.553	2.037	14.02	0.000	24.544	32.562
ITI	logPb	1.684	1.014	1.66	0.098	-0.311	3.679
logN	(Intercept)	4.825	0.141	34.14	0.000	4.547	5.103
logN	logPb	-0.327	0.070	-4.64	0.000	-0.465	-0.188
M_AMBI	(Intercept)	0.620	0.015	40.81	0.000	0.590	0.650
M_AMBI	logPb	-0.046	0.008	-6.08	0.000	-0.061	-0.031
MEDOCC	(Intercept)	1.993	0.127	15.66	0.000	1.742	2.243
MEDOCC	logPb	0.241	0.063	3.80	0.000	0.116	0.366
S	(Intercept)	14.096	0.636	22.16	0.000	12.844	15.348
S	logPb	-1.857	0.317	-5.87	0.000	-2.481	-1.234
TBI	(Intercept)	0.356	0.017	20.90	0.000	0.323	0.390
TBI	logPb	-0.043	0.008	-5.10	0.000	-0.060	-0.027

kable(summ_logPb)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.070	0.066	1.041	20.87	0.000	2	-274	555	566	3.01e+02	278
AMBI_S	0.128	0.124	0.769	40.63	0.000	2	-189	385	396	1.64e+02	278
BENTIX	0.034	0.031	1.604	9.90	0.002	2	-395	797	808	7.15e+02	278
BQI	0.080	0.077	2.432	24.29	0.000	2	-512	1030	1041	1.64e+03	278
ITI	0.010	0.006	21.532	2.76	0.098	2	-1122	2251	2262	1.29e+05	278
logN	0.072	0.069	1.494	21.58	0.000	2	-375	757	768	6.21e+02	278
M_AMBI	0.117	0.114	0.161	37.00	0.000	2	249	-492	-481	7.18e+00	278
MEDOCC	0.049	0.046	1.346	14.45	0.000	2	-346	698	709	5.03e+02	278
S	0.110	0.107	6.725	34.43	0.000	2	-797	1599	1610	1.26e+04	278
TBI	0.086	0.082	0.180	26.02	0.000	2	217	-427	-416	9.04e+00	278

###logCu simple regression#####
simple_logCu<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index7 = lm(value ~ logCu,data = .,weights = sqrt(n)))

coef_logCu <- tidy(simple_logCu, fit_index7,conf.int = T)
summ_logCu <- glance(simple_logCu, fit_index7)
kable(coef_logCu)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.507	0.079	18.991	0.000	1.350	1.663
AMBI	logCu	0.206	0.040	5.119	0.000	0.127	0.285
AMBI_S	(Intercept)	1.587	0.058	27.149	0.000	1.472	1.702
AMBI_S	logCu	0.205	0.030	6.908	0.000	0.146	0.263
BENTIX	(Intercept)	5.042	0.122	41.482	0.000	4.803	5.282
BENTIX	logCu	-0.264	0.062	-4.286	0.000	-0.385	-0.143
BQI	(Intercept)	5.874	0.182	32.336	0.000	5.517	6.232
BQI	logCu	-0.597	0.092	-6.484	0.000	-0.778	-0.416
ITI	(Intercept)	30.715	1.662	18.479	0.000	27.443	33.988
ITI	logCu	0.553	0.842	0.656	0.512	-1.105	2.210
logN	(Intercept)	4.647	0.116	40.190	0.000	4.420	4.875
logN	logCu	-0.246	0.059	-4.206	0.000	-0.362	-0.131
M_AMBI	(Intercept)	0.607	0.012	49.833	0.000	0.583	0.631
M_AMBI	logCu	-0.042	0.006	-6.748	0.000	-0.054	-0.030
MEDOCC	(Intercept)	2.023	0.102	19.783	0.000	1.822	2.224
MEDOCC	logCu	0.240	0.052	4.627	0.000	0.138	0.342
S	(Intercept)	13.433	0.514	26.121	0.000	12.421	14.446
S	logCu	-1.600	0.261	-6.141	0.000	-2.113	-1.087
TBI	(Intercept)	0.349	0.014	25.623	0.000	0.322	0.376
TBI	logCu	-0.042	0.007	-6.070	0.000	-0.055	-0.028

kable(summ_logCu)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.086	0.083	1.032	26.209	0.000	2	-272	550	561	2.96e+02	278
AMBI_S	0.147	0.143	0.760	47.727	0.000	2	-186	379	389	1.61e+02	278
BENTIX	0.062	0.059	1.581	18.367	0.000	2	-391	789	800	6.95e+02	278
BQI	0.131	0.128	2.363	42.045	0.000	2	-504	1014	1025	1.55e+03	278
ITI	0.002	-0.002	21.622	0.431	0.512	2	-1124	2253	2264	1.30e+05	278
logN	0.060	0.056	1.504	17.691	0.000	2	-377	761	772	6.29e+02	278
M_AMBI	0.141	0.138	0.159	45.533	0.000	2	253	-499	-488	6.99e+00	278
MEDOCC	0.071	0.068	1.330	21.406	0.000	2	-343	692	703	4.92e+02	278
S	0.119	0.116	6.690	37.713	0.000	2	-795	1596	1607	1.24e+04	278
TBI	0.117	0.114	0.177	36.851	0.000	2	222	-437	-426	8.73e+00	278

###sqrtZn simple regression####
simple_sqrtZn<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index8 = lm(value ~ sqrtZn,data = .,weights = sqrt(n)))

coef_sqrtZn <- tidy(simple_sqrtZn, fit_index8,conf.int = T)
summ_sqrtZn<- glance(simple_sqrtZn, fit_index8)
kable(coef_sqrtZn)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.460	0.104	14.00	0.000	1.255	1.666
AMBI	sqrtZn	0.063	0.015	4.16	0.000	0.033	0.093
AMBI_S	(Intercept)	1.516	0.077	19.63	0.000	1.364	1.668
AMBI_S	sqrtZn	0.066	0.011	5.93	0.000	0.044	0.088
BENTIX	(Intercept)	5.052	0.160	31.62	0.000	4.737	5.366
BENTIX	sqrtZn	-0.073	0.023	-3.15	0.002	-0.118	-0.027
BQI	(Intercept)	5.870	0.244	24.10	0.000	5.391	6.350
BQI	sqrtZn	-0.161	0.035	-4.56	0.000	-0.230	-0.091
ITI	(Intercept)	29.465	2.150	13.70	0.000	25.232	33.698
ITI	sqrtZn	0.344	0.311	1.10	0.270	-0.269	0.957
logN	(Intercept)	4.590	0.153	30.07	0.000	4.289	4.890
logN	sqrtZn	-0.058	0.022	-2.61	0.010	-0.101	-0.014
M_AMBI	(Intercept)	0.620	0.016	38.48	0.000	0.589	0.652
M_AMBI	sqrtZn	-0.013	0.002	-5.69	0.000	-0.018	-0.009
MEDOCC	(Intercept)	2.006	0.135	14.90	0.000	1.741	2.271
MEDOCC	sqrtZn	0.067	0.019	3.46	0.001	0.029	0.106
S	(Intercept)	13.601	0.684	19.88	0.000	12.255	14.948
S	sqrtZn	-0.459	0.099	-4.63	0.000	-0.654	-0.264
TBI	(Intercept)	0.349	0.018	19.18	0.000	0.313	0.385
TBI	sqrtZn	-0.011	0.003	-4.29	0.000	-0.016	-0.006

kable(summ_sqrtZn)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.059	0.055	1.047	17.32	0.000	2	-276	558	569	3.05e+02	278
AMBI_S	0.112	0.109	0.775	35.17	0.000	2	-192	390	400	1.67e+02	278
BENTIX	0.034	0.031	1.604	9.93	0.002	2	-395	797	808	7.15e+02	278
BQI	0.070	0.066	2.446	20.78	0.000	2	-513	1033	1044	1.66e+03	278
ITI	0.004	0.001	21.591	1.22	0.270	2	-1123	2253	2263	1.30e+05	278
logN	0.024	0.020	1.533	6.81	0.010	2	-383	771	782	6.53e+02	278
M_AMBI	0.104	0.101	0.162	32.42	0.000	2	247	-488	-477	7.28e+00	278
MEDOCC	0.041	0.038	1.352	11.98	0.001	2	-347	701	712	5.08e+02	278
S	0.072	0.068	6.869	21.46	0.000	2	-803	1611	1622	1.31e+04	278
TBI	0.062	0.059	0.183	18.43	0.000	2	213	-420	-409	9.27e+00	278

###logTOC simple regression#####
simple_logTOC<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index9 = lm(value ~ logTOC,data = .,weights = sqrt(n)))

coef_logTOC <- tidy(simple_logTOC, fit_index9,conf.int = T) 
summ_logTOC<- glance(simple_logTOC, fit_index9) 
kable(coef_logTOC)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.808	0.114	15.827	0.000	1.581	2.035
AMBI	logTOC	0.565	0.236	2.395	0.019	0.096	1.033
AMBI_S	(Intercept)	1.785	0.089	20.115	0.000	1.608	1.961
AMBI_S	logTOC	0.758	0.183	4.138	0.000	0.393	1.122
BENTIX	(Intercept)	4.830	0.185	26.071	0.000	4.461	5.198
BENTIX	logTOC	-0.683	0.382	-1.787	0.078	-1.443	0.077
BQI	(Intercept)	5.564	0.309	18.003	0.000	4.949	6.179
BQI	logTOC	-1.914	0.638	-3.001	0.004	-3.182	-0.645
ITI	(Intercept)	35.553	2.491	14.273	0.000	30.599	40.507
ITI	logTOC	-6.299	5.140	-1.226	0.224	-16.522	3.924
logN	(Intercept)	4.369	0.218	20.036	0.000	3.935	4.803
logN	logTOC	-0.308	0.450	-0.685	0.495	-1.203	0.587
M_AMBI	(Intercept)	0.592	0.020	29.183	0.000	0.552	0.632
M_AMBI	logTOC	-0.146	0.042	-3.489	0.001	-0.229	-0.063
MEDOCC	(Intercept)	2.370	0.143	16.535	0.000	2.085	2.655
MEDOCC	logTOC	0.703	0.296	2.376	0.020	0.115	1.291
S	(Intercept)	12.948	0.901	14.374	0.000	11.156	14.739
S	logTOC	-5.490	1.859	-2.954	0.004	-9.187	-1.793
TBI	(Intercept)	0.301	0.020	14.881	0.000	0.261	0.341
TBI	logTOC	-0.100	0.042	-2.398	0.019	-0.183	-0.017

kable(summ_logTOC)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.065	0.053	0.963	5.737	0.019	2	-69.4	145	152	77.04	83
AMBI_S	0.171	0.161	0.748	17.123	0.000	2	-48.0	102	109	46.49	83
BENTIX	0.037	0.025	1.562	3.194	0.078	2	-110.5	227	234	202.61	83
BQI	0.098	0.087	2.607	9.005	0.004	2	-154.0	314	321	563.95	83
ITI	0.018	0.006	21.008	1.502	0.224	2	-331.4	669	676	36631.73	83
logN	0.006	-0.006	1.839	0.469	0.495	2	-124.4	255	262	280.75	83
M_AMBI	0.128	0.117	0.171	12.171	0.001	2	77.5	-149	-142	2.43	83
MEDOCC	0.064	0.052	1.209	5.648	0.020	2	-88.7	183	191	121.27	83
S	0.095	0.084	7.597	8.725	0.004	2	-244.9	496	503	4790.21	83
TBI	0.065	0.054	0.171	5.750	0.019	2	77.8	-150	-142	2.41	83

###sqrtTN simple regression#####
simple_sqrtTN <- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index10 = lm(value ~ sqrtTN ,data = .,weights = sqrt(n)))

coef_sqrtTN  <- tidy(simple_sqrtTN , fit_index10,conf.int = T)
summ_sqrtTN <- glance(simple_sqrtTN , fit_index10)
kable(coef_sqrtTN)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.677	0.049	34.541	0.000	1.582	1.773
AMBI	sqrtTN	0.015	0.002	6.186	0.000	0.010	0.020
AMBI_S	(Intercept)	1.750	0.035	49.973	0.000	1.681	1.819
AMBI_S	sqrtTN	0.015	0.002	8.178	0.000	0.011	0.018
BENTIX	(Intercept)	4.862	0.075	64.700	0.000	4.714	5.010
BENTIX	sqrtTN	-0.020	0.004	-5.098	0.000	-0.027	-0.012
BQI	(Intercept)	5.012	0.128	39.270	0.000	4.760	5.263
BQI	sqrtTN	-0.005	0.007	-0.836	0.404	-0.018	0.007
ITI	(Intercept)	32.522	1.028	31.639	0.000	30.498	34.546
ITI	sqrtTN	-0.009	0.053	-0.164	0.870	-0.112	0.095
logN	(Intercept)	4.224	0.080	52.982	0.000	4.067	4.381
logN	sqrtTN	0.004	0.004	0.901	0.368	-0.004	0.012
M_AMBI	(Intercept)	0.563	0.008	67.726	0.000	0.546	0.579
M_AMBI	sqrtTN	-0.001	0.000	-3.335	0.001	-0.002	-0.001
MEDOCC	(Intercept)	2.216	0.063	34.917	0.000	2.091	2.340
MEDOCC	sqrtTN	0.019	0.003	5.867	0.000	0.013	0.025
S	(Intercept)	11.223	0.365	30.770	0.000	10.505	11.942
S	sqrtTN	-0.012	0.019	-0.621	0.535	-0.048	0.025
TBI	(Intercept)	0.279	0.009	30.406	0.000	0.261	0.297
TBI	sqrtTN	0.000	0.000	-0.132	0.895	-0.001	0.001

kable(summ_sqrtTN)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.127	0.123	0.972	38.271	0.000	2	-241	488	499	2.49e+02	264
AMBI_S	0.202	0.199	0.701	66.874	0.000	2	-154	314	325	1.30e+02	264
BENTIX	0.090	0.086	1.504	25.988	0.000	2	-357	721	731	5.97e+02	264
BQI	0.003	-0.001	2.554	0.699	0.404	2	-498	1002	1013	1.72e+03	264
ITI	0.000	-0.004	20.572	0.027	0.870	2	-1053	2112	2123	1.12e+05	264
logN	0.003	-0.001	1.595	0.813	0.368	2	-373	752	763	6.72e+02	264
M_AMBI	0.040	0.037	0.166	11.121	0.001	2	229	-451	-440	7.30e+00	264
MEDOCC	0.115	0.112	1.270	34.421	0.000	2	-312	631	641	4.26e+02	264
S	0.001	-0.002	7.300	0.385	0.535	2	-778	1561	1572	1.41e+04	264
TBI	0.000	-0.004	0.183	0.017	0.895	2	202	-399	-388	8.88e+00	264

###Type simple regression####
simple_type<- 
  dat_simple %>%
  group_by(index) %>%
  do(fit_index11 = lm(value ~ Region,data = .,weights = sqrt(n)))

coef_type<- tidy(simple_type, fit_index11,conf.int = T)
summ_type <- glance(simple_type, fit_index11)
kable(coef_type)

index	term	estimate	std.error	statistic	p.value	conf.low	conf.high
AMBI	(Intercept)	1.884	0.090	20.952	0.000	1.707	2.061
AMBI	RegionEastern North Island	-0.024	0.134	-0.178	0.858	-0.288	0.240
AMBI	RegionNorth Eastern	0.140	0.101	1.387	0.166	-0.058	0.338
AMBI	RegionSouthern	0.305	0.114	2.669	0.008	0.081	0.530
AMBI	RegionWestern North Island	0.061	0.120	0.510	0.610	-0.174	0.296
AMBI_S	(Intercept)	1.947	0.071	27.380	0.000	1.807	2.087
AMBI_S	RegionEastern North Island	0.048	0.106	0.447	0.655	-0.161	0.256
AMBI_S	RegionNorth Eastern	0.130	0.080	1.625	0.105	-0.027	0.286
AMBI_S	RegionSouthern	0.253	0.090	2.798	0.005	0.075	0.431
AMBI_S	RegionWestern North Island	-0.041	0.095	-0.437	0.663	-0.227	0.145
BENTIX	(Intercept)	4.547	0.133	34.310	0.000	4.287	4.807
BENTIX	RegionEastern North Island	0.405	0.198	2.044	0.041	0.016	0.794
BENTIX	RegionNorth Eastern	-0.226	0.149	-1.522	0.129	-0.518	0.066
BENTIX	RegionSouthern	-0.437	0.169	-2.596	0.010	-0.768	-0.106
BENTIX	RegionWestern North Island	-0.412	0.176	-2.335	0.020	-0.758	-0.065
BQI	(Intercept)	5.128	0.166	30.934	0.000	4.803	5.454
BQI	RegionEastern North Island	-1.788	0.248	-7.220	0.000	-2.275	-1.302
BQI	RegionNorth Eastern	0.427	0.186	2.298	0.022	0.062	0.792
BQI	RegionSouthern	-0.815	0.211	-3.868	0.000	-1.229	-0.401
BQI	RegionWestern North Island	0.975	0.221	4.422	0.000	0.542	1.408
ITI	(Intercept)	29.885	1.646	18.155	0.000	26.651	33.119
ITI	RegionEastern North Island	13.064	2.460	5.312	0.000	8.233	17.896
ITI	RegionNorth Eastern	-0.916	1.845	-0.496	0.620	-4.541	2.709
ITI	RegionSouthern	-0.303	2.093	-0.145	0.885	-4.414	3.809
ITI	RegionWestern North Island	-5.897	2.190	-2.693	0.007	-10.199	-1.596
logN	(Intercept)	3.993	0.094	42.648	0.000	3.809	4.177
logN	RegionEastern North Island	-0.755	0.140	-5.394	0.000	-1.029	-0.480
logN	RegionNorth Eastern	0.504	0.105	4.800	0.000	0.298	0.710
logN	RegionSouthern	0.514	0.119	4.317	0.000	0.280	0.748
logN	RegionWestern North Island	0.334	0.125	2.679	0.008	0.089	0.578
M_AMBI	(Intercept)	0.554	0.013	41.533	0.000	0.528	0.580
M_AMBI	RegionEastern North Island	-0.085	0.020	-4.287	0.000	-0.125	-0.046
M_AMBI	RegionNorth Eastern	0.011	0.015	0.731	0.465	-0.018	0.040
M_AMBI	RegionSouthern	-0.067	0.017	-3.940	0.000	-0.100	-0.034
M_AMBI	RegionWestern North Island	0.040	0.018	2.276	0.023	0.006	0.075
MEDOCC	(Intercept)	2.488	0.115	21.674	0.000	2.263	2.714
MEDOCC	RegionEastern North Island	-0.141	0.172	-0.824	0.411	-0.478	0.196
MEDOCC	RegionNorth Eastern	0.165	0.129	1.280	0.201	-0.088	0.417
MEDOCC	RegionSouthern	0.398	0.146	2.726	0.007	0.111	0.685
MEDOCC	RegionWestern North Island	0.089	0.153	0.585	0.559	-0.211	0.389
S	(Intercept)	11.085	0.496	22.366	0.000	10.111	12.059
S	RegionEastern North Island	-4.596	0.741	-6.206	0.000	-6.051	-3.141
S	RegionNorth Eastern	1.410	0.556	2.538	0.011	0.319	2.502
S	RegionSouthern	-1.340	0.630	-2.126	0.034	-2.578	-0.102
S	RegionWestern North Island	2.173	0.659	3.295	0.001	0.878	3.468
TBI	(Intercept)	0.302	0.014	22.167	0.000	0.275	0.328
TBI	RegionEastern North Island	-0.138	0.020	-6.806	0.000	-0.178	-0.098
TBI	RegionNorth Eastern	0.049	0.015	3.188	0.002	0.019	0.079
TBI	RegionSouthern	-0.072	0.017	-4.189	0.000	-0.106	-0.039
TBI	RegionWestern North Island	0.058	0.018	3.203	0.001	0.022	0.094

kable(summ_type)

index	r.squared	adj.r.squared	sigma	statistic	p.value	df	logLik	AIC	BIC	deviance	df.residual
AMBI	0.021	0.014	1.189	2.86	0.023	5	-588	1189	1214	754.5	534
AMBI_S	0.029	0.021	0.940	3.92	0.004	5	-462	936	962	472.0	534
BENTIX	0.050	0.042	1.752	6.95	0.000	5	-797	1607	1633	1639.0	534
BQI	0.277	0.271	2.192	51.09	0.000	5	-918	1848	1874	2564.6	534
ITI	0.114	0.108	21.760	17.25	0.000	5	-2155	4323	4348	252856.9	534
logN	0.208	0.202	1.238	35.07	0.000	5	-610	1232	1258	817.9	534
M_AMBI	0.136	0.130	0.176	21.02	0.000	5	440	-868	-842	16.6	534
MEDOCC	0.027	0.020	1.517	3.73	0.005	5	-720	1452	1478	1229.7	534
S	0.202	0.196	6.552	33.84	0.000	5	-1508	3029	3054	22922.3	534
TBI	0.275	0.269	0.180	50.62	0.000	5	429	-847	-821	17.3	534

simple_regressions.R

javiera

Wed May 24 10:49:47 2017