Stroke Prediction Analysis

• Dataset taken from Kaggle (posted by: fedesoriano)
• Categorical Variables
  • gender; ever_married
  • hypertension; heart_disease
  • work_type; residence_status
  • smoking status
• Numeric Variables
  • age
  • avg_glucose_level
  • bmi
• Target Variable: stroke (Heart Stroke? Binary)
  

• Quick “Trial and Error” on predictors
• Shows the coefficients with their p-values (significance)
• Results Sorted in increasing order of p-values
• Significant coefficients highlighted with asterix

• Choice of Scatter plot Numeric Variables (X and Y)
• “Third Dimension” option for Color on a Categorical Variable
• Target Variable (stroke) always indicated by Size of Bubble
• Different Options of Color Palettes available to accommodate for accessibility

library(ggplot2)
source(paste0(getwd(),"/StrokePrediction/getDF.R"))
summary(glm(stroke~age,data=df.train,family="binomial"))$coef

             Estimate  Std. Error   z value     Pr(>|z|)
(Intercept) -7.026230 0.360701909 -19.47933 1.644287e-84
age          0.071733 0.005337112  13.44041 3.504734e-41