House-Mortgage-Dataset-Analysis

Part-1 Exploratory Dataset Analysis Home Mortgage NY

 - Loading and analyzing dataset
 - Visualizing the dataset
 - Building a classification model to decide which predictors are most important
 - Calculating the accuracy of the model by plotting ROC curve

Part-2

The main aim of the project was to predict whether a mortgage application will be accepted or not. Dataset was extracted from AER package in R.

Steps involved:

 --Data extraction and cleansing
 --Outliers detection using QQ-Plot and box-plots
 --Skewness detection using density plots
 --Model fitting using GLM
 --Model selection using forward and hybrid methods
 --Choosing the best model using ANOVA
 --Predicting the model accuracy 
 --Plotting the ROC curve 
 --Conclusion:
   On the basis of various analysis performed; the most significant predictors are:
   Payment to Income Ratio
   Loan to value ratio  
   Credit history: consumer payments
   Public bad credit record
   Insurance
   Ethnicity
   Marital status