README.md

In RashiMohta/COVID-19-cases-prediction: Jump-drop adjusted prediction of cumulative infected cases using the modified SIS model

COVID-19-cases-prediction

JUMP-DROP-ADJUSTED PREDICTION OF CUMULATIVE CASES OF COVID-19 USING MODIFIED SIS MODEL

Installation

This library can be intatalled using the following command: library(devtools) install_github("RashiMohta/adjustedPredSIS")

This library can be imported using: library(adjustedPred)

Additional libraries to be imported: library(pracma) library(ggplot2)

Define constants

name_of_state = "DL" ub_for_adjustment = 5 bound_metric = "C3_1day" population = 18710922 gamma = 1 / 14.0 start_date = "2020-3-13" cur_date = "2021-4-12" cur_day = as.numeric(difftime(cur_date, start_date, units = "days")) cur_day = 400 last_n_day = 20 last_limit = 30 next_n_days = 20

Read data

file_name <- "/home/rashi/BTP/btp/state_wise_daily.csv" data <- read.csv(file_name) state_data = data[, c("Status", "Date", name_of_state)]

Format the date according to the dataset

for (i in 1:nrow(state_data)) { date <- state_data$Date[i] lst <- unlist(strsplit(as.character(date) , '-')) if (lst[2] == "Sept") state_data[i, 2] <- strcat(strcat(lst[1], "Sep", '-'), lst[3], '-') } colnames(state_data) <- c("Status", "Date", "Count") date <- vector() confirmed_values <- vector() adjusted_confirmed_values <- vector()

for (i in 1:nrow(state_data)) { row <- state_data[i,] status <- row$Status if( status == "Confirmed" ){ date <- append(date,toString(row$Date)) confirmed_values <- append(confirmed_values, abs(row$Count) ) #store original values } }

Obtain results

methods = c("mean", "linear interpolation", "end points mean", "percentile");

for (method in methods){ df_confirmed_values <- data.frame(date,confirmed_values) result = compare_results(data=state_data, bound_metric='C3_1day',df_confirmed_values=df_confirmed_values, last_n_day=20, next_n_days=20, cur_date="2020-10-19", method = method ) }

RashiMohta/COVID-19-cases-prediction documentation built on Oct. 26, 2024, 9:48 a.m.