PACOTES UTILIZADOS:
library(lavaan)
## This is lavaan 0.6-7
## lavaan is BETA software! Please report any bugs.
library(semPlot)
## Registered S3 methods overwritten by 'huge':
## method from
## plot.sim BDgraph
## print.sim BDgraph
library(OpenMx)
## To take full advantage of multiple cores, use:
## mxOption(key='Number of Threads', value=parallel::detectCores()) #now
## Sys.setenv(OMP_NUM_THREADS=parallel::detectCores()) #before library(OpenMx)
##
## Attaching package: 'OpenMx'
## The following object is masked from 'package:lavaan':
##
## vech
library(tidyverse)
## ── Attaching packages ───────────────────────────────────────────────────────────── tidyverse 1.3.0 ──
## ✓ ggplot2 3.3.2 ✓ purrr 0.3.4
## ✓ tibble 3.0.3 ✓ dplyr 1.0.2
## ✓ tidyr 1.1.2 ✓ stringr 1.4.0
## ✓ readr 1.3.1 ✓ forcats 0.5.0
## ── Conflicts ──────────────────────────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
library(knitr)
library(kableExtra)
##
## Attaching package: 'kableExtra'
## The following object is masked from 'package:dplyr':
##
## group_rows
library(GGally)
## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2
LEITURA DE DADOS GERAIS (CANA PLANTA E SOCA):
DADOS = read.csv("DADOS_CANA.csv", header = TRUE, sep = ";", dec = ",")
str(DADOS)
## 'data.frame': 108 obs. of 7 variables:
## $ Clones: chr "RB975375" "RB975375" "RB975375" "RB006655 " ...
## $ Rep : int 1 2 3 1 2 3 1 2 3 1 ...
## $ FIBRA : num 12.7 12.6 12.8 13.3 13 ...
## $ PZA : num 85.4 84.8 85.9 83.9 81.1 ...
## $ PCC : num 15.7 15.7 15.7 14.9 13.5 ...
## $ ATR : num 155 154 154 148 137 ...
## $ TIPO : chr "PLANTA" "PLANTA" "PLANTA" "PLANTA" ...
DADOS$Clones = as.factor(DADOS$Clones)
DADOS$Rep = as.factor(DADOS$Rep)
DADOS$TIPO = as.factor(DADOS$TIPO)
dados_planta = DADOS[which(DADOS$TIPO=="PLANTA"),]
dados_soca = DADOS[which(DADOS$TIPO=="SOCA"),]
ANALISES DE VARIÂNCIA INDIVIDUAIS E CONJUNTA
Individuais:
lm_planta = lm(ATR~Clones+Rep+FIBRA+PZA+PCC, data = dados_planta)
anova(lm_planta)
## Analysis of Variance Table
##
## Response: ATR
## Df Sum Sq Mean Sq F value Pr(>F)
## Clones 17 4101.6 241.268 29.7623 1.481e-14 ***
## Rep 2 6.2 3.115 0.3843 0.68412
## FIBRA 1 51.8 51.834 6.3941 0.01676 *
## PZA 1 176.5 176.510 21.7739 5.578e-05 ***
## PCC 1 42.7 42.662 5.2627 0.02872 *
## Residuals 31 251.3 8.107
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
lm_soca = lm(ATR~Clones+Rep+FIBRA+PZA+PCC, data = dados_soca)
anova(lm_soca)
## Analysis of Variance Table
##
## Response: ATR
## Df Sum Sq Mean Sq F value Pr(>F)
## Clones 17 6618.8 389.34 30.8398 8.933e-15 ***
## Rep 2 2.5 1.27 0.1008 0.9044
## FIBRA 1 1.4 1.43 0.1129 0.7391
## PZA 1 34.1 34.09 2.7000 0.1105
## PCC 1 500.5 500.54 39.6480 5.290e-07 ***
## Residuals 31 391.4 12.62
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Conjunta:
lm_geral = lm(ATR~Clones+Rep%in%TIPO+FIBRA+PZA+PCC+TIPO+Clones:TIPO, data = DADOS)
anova(lm_geral)
## Analysis of Variance Table
##
## Response: ATR
## Df Sum Sq Mean Sq F value Pr(>F)
## Clones 17 9605.0 565.00 47.5694 < 2.2e-16 ***
## FIBRA 1 81.8 81.84 6.8903 0.01079 *
## PZA 1 938.3 938.34 79.0017 7.794e-13 ***
## PCC 1 889.6 889.63 74.9006 2.016e-12 ***
## TIPO 1 11.8 11.78 0.9919 0.32297
## Rep:TIPO 4 14.7 3.67 0.3090 0.87098
## Clones:TIPO 17 368.7 21.69 1.8259 0.04330 *
## Residuals 65 772.0 11.88
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
De acordo com o modelo testado, existem efeitos acentuados de todas as variáveis, bem como de clones na resposta de ATR. Não existem diferenças claras da influência do tipo de plantio (cana ou soca) sob ATR, no entanto foi observada interação entre o efeito de clones e o plantio.
ANÁLISE DE TRILHA (planta e soca)
modelo = 'ATR~FIBRA+PZA+PCC'
fit = cfa(modelo, data = DADOS)
summary(fit,fit.measures = TRUE, standardized=T,rsquare=T)
## lavaan 0.6-7 ended normally after 25 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 4
##
## Number of observations 108
##
## Model Test User Model:
##
## Test statistic 0.000
## Degrees of freedom 0
##
## Model Test Baseline Model:
##
## Test statistic 196.737
## Degrees of freedom 3
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.000
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -312.231
## Loglikelihood unrestricted model (H1) -312.231
##
## Akaike (AIC) 632.461
## Bayesian (BIC) 643.190
## Sample-size adjusted Bayesian (BIC) 630.551
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value RMSEA <= 0.05 NA
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## ATR ~
## FIBRA -0.258 0.539 -0.480 0.632 -0.258 -0.021
## PZA -0.044 0.149 -0.294 0.769 -0.044 -0.015
## PCC 8.405 0.501 16.765 0.000 8.405 0.915
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .ATR 18.995 2.585 7.348 0.000 18.995 0.162
##
## R-Square:
## Estimate
## ATR 0.838
semPaths(fit,"std",layout = 'tree', edge.label.cex=.9, curvePivot = TRUE)
Correlação de Pearson:
ggcorr(DADOS[-c(1, 2,7)], nbreaks = 6, label = T, low = "red3", high = "green3",
label_round = 2, name = "Correlation Scale", label_alpha = T, hjust = 0.75) +
ggtitle(label = "Correlation Plot") +
theme(plot.title = element_text(hjust = 0.6))
Analise de trilha: cana planta
modelo_planta = 'ATR~FIBRA+PZA+PCC'
fit_planta = cfa(modelo_planta, data = dados_planta)
summary(fit_planta,fit.measures = TRUE, standardized=T,rsquare=T)
## lavaan 0.6-7 ended normally after 22 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 4
##
## Number of observations 54
##
## Model Test User Model:
##
## Test statistic 0.000
## Degrees of freedom 0
##
## Model Test Baseline Model:
##
## Test statistic 92.667
## Degrees of freedom 3
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.000
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -150.476
## Loglikelihood unrestricted model (H1) -150.476
##
## Akaike (AIC) 308.952
## Bayesian (BIC) 316.908
## Sample-size adjusted Bayesian (BIC) 304.341
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value RMSEA <= 0.05 NA
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## ATR ~
## FIBRA -0.995 0.575 -1.731 0.084 -0.995 -0.113
## PZA -0.111 0.167 -0.664 0.506 -0.111 -0.048
## PCC 7.148 0.623 11.473 0.000 7.148 0.884
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .ATR 15.414 2.967 5.196 0.000 15.414 0.180
##
## R-Square:
## Estimate
## ATR 0.820
semPaths(fit_planta, 'std', layout = 'circle')
semPaths(fit_planta,"std",layout = 'tree', edge.label.cex=.9, curvePivot = TRUE)
Analise de trilha: cana soca
modelo_soca = 'ATR~FIBRA+PZA+PCC'
fit_soca = cfa(modelo_soca, data = dados_soca)
summary(fit_soca,fit.measures = TRUE, standardized=T,rsquare=T)
## lavaan 0.6-7 ended normally after 29 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of free parameters 4
##
## Number of observations 54
##
## Model Test User Model:
##
## Test statistic 0.000
## Degrees of freedom 0
##
## Model Test Baseline Model:
##
## Test statistic 121.689
## Degrees of freedom 3
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.000
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -149.162
## Loglikelihood unrestricted model (H1) -149.162
##
## Akaike (AIC) 306.325
## Bayesian (BIC) 314.281
## Sample-size adjusted Bayesian (BIC) 301.714
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value RMSEA <= 0.05 NA
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.000
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## ATR ~
## FIBRA 0.923 1.023 0.902 0.367 0.923 0.044
## PZA 0.083 0.232 0.357 0.721 0.083 0.022
## PCC 10.130 0.687 14.753 0.000 10.130 0.949
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .ATR 14.683 2.826 5.196 0.000 14.683 0.105
##
## R-Square:
## Estimate
## ATR 0.895
semPaths(fit_soca, 'std', layout = 'circle')
semPaths(fit_soca,"std",layout = 'tree', edge.label.cex=.9, curvePivot = TRUE)
De forma geral PCC é um indicador importante de ATR (p<0,01 em todos os casos). PZA não exerce influência direta em ATR, no entanto está positivamente relacionado a PCC. FIB apresenta relação negativa com PCC e também não influencia ATR diretamente.