gravatar

billa

Manoj Kumar Billa

Recently Published

PGA (Professional Golfers Association) Data Analysis
PGA data is taken from sportsillustrated.cnn.com and the performance statistics and winnings for 196 PGA participants during 2004 season is analysed using Linear Regression Analysis.
Wine Quality Analysis
Analysis of 1599 varieties of Wines resulting in a linear regression model relating the various physio chemical components of red-wine to the wine quality ratings by wine tasters. Data is checked for null values, normality and outliers. After visualizing correlations using scatter plots, a linear model is fit using the 'Backward Elimination' regression method. After arriving at an optimum equation, the ANOVA method is used to eliminate some more insignificant variables. After this, model adequacy tests are done by checking residual plots and predicted vs fitted plots. Transformation is done based on observed plots. After this, cooks distance is calculated to check for outliers once again and data is checked for multi-collinearity. Variables are centered and final Linear regression model is arrived upon.