MartinMacharia

Model selection codes
Model 1
Initial model specification
library(lme4)
Model1=lmer(yield.per.ha~n.per.ha*rain*soilclass+manure+seed+n.squared+(1|year)+(1|hhid)+(1|aezsmall)+(1|aez),data=isfm)
Model selection using fitLMER.fnc
library(LMERConvenienceFunctions)
optimum1=fitLMER.fnc(Model1,backfit.on="F",item=F,alpha=0.05,if.warn.not.add=TRUE,llrt=T,prune.ranefs=TRUE,p.value="upper",t.threshold=2,set.REML.FALSE=TRUE,reset.REML.TRUE=TRUE)
optimum1
Model criticism plots

MartinMacharia

Guide to Generating field book and field layouts in R
January 2014 

What is R?
R is a programming language for statistical computing and graphics. It is available as free software. As a thriving open-source project, R is supported by a large community of users and developers worldwide. Whether you're using R for data management, analyzing genomic sequences, or designing field books and layouts for agronomy work, experts in every domain have made resources, applications and code available for free, online. 
Why we need R in OFRA
R presents a quick way to randomize treatments and generate field books and field layouts. Beyond this R is the leading tool for statistics, data analysis, and machine learning.  It is more than a statistical package; it’s a programming language, so you can create your own objects, functions, and packages.
What is a package?
Peer reviewed methods in statistics and statistical modeling from lead researchers in statistics that have been packaged into applications and can be downloaded f

MartinMacharia

#Appendix 2: Optimization algorithm  
#Specify the function to be maximized. By default constrOptim performs minimization but the problem can be maximized by setting the fnscale to -1
LP_sol=function(X){
  Y_R=5625-2897*(0.966^X[1])-2728   #Rice N response function
  Y_M=3159-1191*(0.976^X[2])-1968   #Maize N response function
  Y_B=1415-715*(0.950^X[3])-700     #Bean N response function
  Y_FM=2100-923*(0.944^X[4])-1177   #Finger millet N response 
  Y_R1=5665-828*(0.871^X[5])-4837   #Rice P response function
  Y_M1=4474-770*(0.898^X[6])-3704   #Maize P response function
  Y_B1=1138-263*(0.848^X[7])-875    #Bean P response function

MartinMacharia

# 1,000,000
Y_R=5625-2897*(0.966^120)-2728   #Rice N response function
Y_M=3159-1191*(0.976^120)-1968   #Maize N response function
Y_B=1415-715*(0.950^68)-700     #Bean N response function
Y_FM=2100-923*(0.944^61)-1177    #Finger millet N response function
Y_R1=5665-828*(0.871^30)-4837     #Rice P response function
Y_M1=4474-770*(0.898^36)-3704     #Maize P response function
Y_B1=1138-263*(0.848^22)-875      #Bean P response function
Y_FM1=2101-537*(0.798^19)-1564    #Finger millet P response function
Y_PP=2538-487*(0.758^19)-487     #Pigeon Pea P response function

MartinMacharia

# soils
# set this to the xlsx file on your computer
setwd("c:/Users/machariam/Desktop/Soil_R_project") # Change working directory
# load libraries
lapply(c("data.table", "DescTools", "stringr", "ggplot2", "readxl", "ranger", "RWeka", "Boruta", "DMwR"), require, character.only = T)
# load data
myfiles = dir(pattern = "xlsx") #These functions produce a character vector of the names of files or directories in the named directory.
myfiles #See no value in this unless it can be linked with the read_excel below
df = read_excel(myfiles, na = "") # Load the excel, still need a way to sort out the data structure and missing values
df

MartinMacharia

#Specify the function to be maximized. By default constrOptim performs minimization but 
#the problem can be maximized by setting the fnscale to -1
LP_sol=function(X){
  Y_R=5625-2897*(0.966^X[1])-2728   #Rice N response function
  Y_M=3159-1191*(0.976^X[2])-1968   #Maize N response function
  Y_B=1415-715*(0.950^X[3])-700     #Bean N response function
  Y_FM=2100-923*(0.944^X[4])-1177    #Finger millet N response function
  Y_R1=5665-828*(0.871^X[5])-4837     #Rice P response function
  Y_M1=4474-770*(0.898^X[6])-3704     #Maize P response function
  Y_B1=1138-263*(0.848^X[7])-875      #Bean P response function

MartinMacharia

#One dimension problem
fun=function(X){
Y=5151-1735*(0.945^X[1])-3416
Py=100
R=Y*Py
Px=200
C=X[1]*Px 
P=Y*Py-X[1]*Px
return(P)
#plot(X,P,type="l")

MartinMacharia

# Spillman/ Assymptotic
#Niger, 2015
#Dosso
#Bengou
prate=c(0,7.5,15,22.5)
yield=c(826.5,1127.4, 1203.7, 1250)
#prate<- c(0,7.5,15,22.5, 30, 32.5)
#yield <- c(826.5,1035.4,1203.7,1250, 1250, 1250)
dat <- as.data.frame(cbind(prate,yield))
dat

MartinMacharia

#Regression kriging for maize 50 kg N

library(rgdal)
library (raster)
library (maptools)
library(rgeos)
library(dismo)
library(sp)

#load tiffs of standardized variables and PCA

MartinMacharia

yield=read.csv(file.choose(),header=T)
yield
#Another way of importing the data
#SmallYieldData <- read.csv("C:/Users/Administrator/Desktop/SmallYieldData.csv")
#View(SmallYieldData)
#You can also type the data in R as follows
yield=data.frame(Nrates=c(0,10,20,30,40,50,60),yield=c(2748,3162,3307,3571,3576,3544,3600))
yield
#You can view in which directory you are in R and also change the working direcory
#as follows