Supplementary material_compare approaches.R

# Test if the array approach and cohort trace approach give similar results. 

################################################################################
# This code forms the basis for: 
# Krijkamp EM, Alarid-Escudero F, Enns EA, Pechlivanoglou P, 
# Hunink MGM, Yang, A., Jalal HJ.
# A multidimensional array representation of state-transition model dynamics. 
# Med dec making(revised, September 2019)
# Please cite the article when using this code
################################################################################

# Run the array approach
source("Supplementary material_array_approach_R code of the stylistic 3-state model.R")

# Store the array approach values in new variables
v_results_array_approach <- v_results
m_M_array_approach <- m_M

# Run the cohort trace approach
source("Supplementary material_cohort_trace_approach_R code of the stylistic 3-state model.R")

# Store the results in a new variable
v_results_cohort_trace_approach <- v_results

# Create a new trace matrix to sum the Markov trace of the cohort trace approach.
m_M_cohort_trace_approach <- matrix(NA, 
                                    nrow = n_t + 1 , 
                                    ncol  = 3)
colnames(m_M_cohort_trace_approach) <- colnames(m_M_array_approach) # name the columns

# Store the data in the right column
m_M_cohort_trace_approach[, "H"] <- m_M[, "H"]
m_M_cohort_trace_approach[, "S"] <- m_M[,"Stemp"] + m_M[, "S"] # sum the proportions of sick (Stemp + S)
m_M_cohort_trace_approach[, "D"] <- m_M[,"Dtemp"] + m_M[, "D"] # sum the proportions of dead (Dtemp + D)

# Compare the two Markov traces rounded on 10 decimal points
round(m_M_cohort_trace_approach, 10) == round(m_M_array_approach, 10)
# TRUE indicated that the values are identical 

# Compare the costs and effects rounded on 5 decimal points
round(v_results_array_approach["Costs"], 10) == round(v_results_cohort_trace_approach["Costs"], 10)
round(v_results_array_approach["Effect"], 10) == round(v_results_cohort_trace_approach["Effect"], 10)