data-550-group1-midterm

Group 1's Midterm Project for DATA 550

---

Objective

In this project, we will estimate the effect of chronic conditions (specifically: diabetes, COPD, asthma, immune suppression, CVD, and obesity) on hospitalization and death of COVID-19 patients in Mexico.

Source: https://datos.gob.mx/busca/dataset/informacion-referente-a-casos-covid-19-en-mexico

---

Team Members

• Pragati Prasad
• Jess Chan
• Kanak Belgaum
• Ariana Parquette
• Sim Fan

---

Code Description

code/01_data_cleaning.R

• run with make data
• inputs raw data from raw_data/ folder
  • raw_data/covid_sub.csv
• Handle missing data
• Set outcome variables to binary
• Change dependent variables (chronic conditions) to either binary or categorical with reference groups
• outputs cleaned/final dataset to output/ folder
  • output/data_cleaned.rds

code/02_descriptive_stat.R

• run with make descriptive
• inputs cleaned dataset from output/ folder
  • output/data_cleaned.rds
• Make table 1 describing population data
• Histograms of dependent variables (chronic conditions)
• Scatter plot to visualize distribution of dependent variables (chronic conditions) amongst the patient data
• outputs all plots and tables to output/ folder
  • output/02_table_one.rds
  • output/02_conditioncount.png
  • output/02_table2.png
  • output/02_heatmap.png

code/03_modeling.R

• run with make regression
• inputs cleaned dataset from output/ folder
  • output/data_cleaned.rds
• runs logistic regression models to estimate the effect of chronic conditions on hospitalization and death of COVID-19 patients in Mexico
• outputs regression models in output folder
  • output/model_1.rds
  • output/model_2.rds
  • output/model_1_results.rds
  • output/model_2_results.rds

code/04_visualization.R

• run with make visualize
• inputs regression files from output/ folder
  • output/data_cleaned.rds
  • output/model_1.rds
  • output/model_2.rds
• Create regression model table
• Diagnostic plots to evaluate model selection
  • Assess linearity assumption
• outputs diagnostic plots to output/ folder
  • output/model_1_effects.rds
  • output/model_2_effects.rds

report.Rmd

• run with make report.html
• inputs all plots and tables
• outputs formatted report file