Assignment 8 PDF

Preview

Summary

Assignment 8...

Description

Assignment 8

file:///C:/Users/Sa3doun/OneDrive - Queen's University/4th year Uni/BIOL343/Assignment 8/..

Assignment 8 library(car) library(ggplot2) library(ggfortify) library(dplyr) library(emmeans) library(MuMIn) options(na.action = “na.fail”) 1. levels_history = 2 levels_bmi = 3 n_per_trt = 30 g_mean = 61 sres = 20 total_n = levels_bmi * levels_history * n_per_trt (ns = matrix(c(rep(as.integer(total_n/(levels_historylevels_bmi)), levels_historylevels_bmi)),nrow = 2)) hist.effect = c(5,-5) bmi.effect = c(-5,-2, 7) hxb.effect = matrix(c(-2,-3,0,0,3,2), nrow = 2) 2. Ward, M.A., Carlsson, C.M., Trivedi, M.A. et al. The effect of body mass index on global brain volume in middle-aged adults: a cross sectional study. BMC Neurol 5, 23 (2005). https://doi.org/10.1186/1471-2377-5-23 (https://doi.org/10.1186/1471-2377-5-23) 3. history = vector(‘numeric’) bmi = vector(‘numeric’) count = 1 for(i in 1:levels_history){ for(j in 1:levels_bmi){ for(k in 1:ns[i,j]){ history[count] = i bmi[count] = j count = count + 1 } } } dat = data.frame(history = factor(history), bmi = factor(bmi)) 4. for(i in 1:length(dat history)){ dat resp = rep(NA, length(dat dat sres) }}

resp)){ history[j]] + bmi.effect[dat$bmi[j]] + hxb.effect[dat

bmi[j]], sd =

5. nreps = 1000 p_hist = rep(NA,nreps) p_bmi = rep(NA,nreps) p_hxb = rep(NA,nreps) for(i in 1:nreps){ for(j in 1:length(dat resp)){ dat resp[j] = rnorm(1, mean = g_mean + hist.effect[dat$history[j]] + bmi.effect[dat$bmi[j]] + hxb.effect[dat bmi[j]], sd = sres) }} 6 sres_extra = c(5,10,15,20,25,30,35,40,45,50) hist_power = rep(NA, length(sres_extra)) bmi_power = rep(NA, length(sres_extra)) hxb_power = rep(NA, length(sres_extra))

1 of 2

2021-01-11, 3:40 p.m.

Assignment 8

file:///C:/Users/Sa3doun/OneDrive - Queen's University/4th year Uni/BIOL343/Assignment 8/..

dat1 = dat dat1

resp))

for(i in 1:length(sres_extra)){ p_hist = rep(NA, nreps) p_bmi =rep(NA, nreps) p_hxb = rep(NA, nreps) for(k in 1:nreps){ dat = dat1 for (j in 1:length(dat resp)){ dat resp[j] = rnorm(1, mean = g_mean + hist.effect[dat$history[j]] + bmi.effect[dat$bmi[j]] + hxb.effect[dat bmi[j]], sd = sres_extra[i]) } res = Anova(lm(resp ~ history * bmi, data = dat)) p_hist[k] = as.numeric(res “Pr(>F)”[2]) p_hxb[k] = as.numeric(res$“Pr(>F)”[3]) } hist_sig = ifelse(p_hist...