Conducted ANOVA and Logistic regression analysis.
The main aim of this project is to perform One-Way ANOVA analysis on the given set of data(values in various levels of education) using python. We build a model that outputs the summary and gives anova table. We set hypothesis for the given data and calculate F-statistic. From F-statistic, p-value is calculated. If the p-value is less than significance level, we reject Null hypothesis which refers to that means of all groups are not equal and the observed difference in the means is not due to sampling variability. After performing hypothesis test, we perform multiple pairwise comparisons of different groups using t-test to determine which means are different. In conclusion, we determine whether the mean of various levels of education is same or which levels of education have different means.
The main aim of this project is to perform logistic regression analysis on the given data set that represents whether a given e-mail is spam or not spam. The dataset contains 20 features that are used to determine whether an e-mail is spam or not spam. Before performing logistic regression, we perform feature elimination so that significant feature sets are used in model analysis. After modeling the data, we iterate the model for various threshold probability values and check the values of sensitivity and specificity for various thresholds.
Therefore, our goal is to find the optimal threshold value for which the true positive rate is close to 1 so that we build an optimum classification model that classifies a spam e-mail from ham.
- Abstract
- Theory
- Exploratory Data analysis
- Analysis Results & Explanation
- Conclusion