Apply Data Analysis Course
Tools / Frameworks: Using R
Duration: 2 Weeks
Cost: $100
MODULE 1
VISUALIZATION WITH GGPLOT2
- Grammar of graphics concept
- Structure of ggplot: data, aes(),
geoms, themes - Bar charts, histograms, density plots
- Boxplots, violin plots
- Scatterplots and trend lines
- Faceting and multi-panel plots
- Customizing titles, labels, legends
- Saving and exporting plots
- Using visualization for exploratory
data analysis (EDA)
MODULE 2
MORE R CODING
- Review of data types and objects
- dplyr verbs: filter, select, mutate,
arrange, summarise, group_by - Using pipes (%>%)
- Handling missing values (NA)
- Conditional statements (ifelse)
- Creating functions
- Loops and apply family
- Data reshaping: pivot_longer,
pivot_wider - Importing and exporting data
- Writing reproducible code
MODULE 3
CORRELATION
- Meaning of correlation
- Positive, negative and zero correlation
- Pearson and Spearman correlation
- Correlation matrix
- Visualization using scatterplots and
heatmaps - Testing significance
- Interpretation and limitations
- Correlation vs causation
MODULE 4
LINEAR REGRESSION
- Concept of linear regression
- Model equation
- Running regression using lm()
- Interpreting coefficients
- Residual analysis
- R-squared and Adjusted Rsquared
- Checking model assumptions
- Plotting regression lines Making
predictions
MODULE 5
MULTIPLE REGRESSION
- Why multiple regression
- Multiple predictors in a model
- Handling categorical variables
- Interpreting partial effects
- Multicollinearity and VIF
- Interaction effects
- Model diagnostics
- Prediction using multiple
regression
Module 6
Logistic Regression
- When to use logistic regression
- Binary outcome variables
- Logistic function and odds Using
glm() - Interpreting odds ratios
- Predicted probabilities
- Confusion matrix
- Accuracy, sensitivity, specificity
- ROC curve (introduction)