Syllabus

General Information

Term/Year: 2025/2026
Subject Code and Course Number: [#856; #867]
Course Title: Data Analytics
Instructor: Karen Hovhannisyan
Email: khovhannisyan91@gmail.com
Phone: +374 77 600 945
Start date: 18-12-2025
End date: 01-05-2026
Class Schedule:
- Tuesdays 19:30–22:00, Saturdays 15:00–17:30 (Cohort #856)
- Thursdays 18:30–21:00, Sundays 10:00–12:30 (Cohort #867)
Materials: Reading Materials
Technical stack: excel, PostgreSQL, Python (Data Analytics toolpack), Tableau, Docker, Git
Prerequisites: Curiosity; Analytical Thinking

Schedule & Topics

Introduction to Data Analytics & Statistical Thinking

Duration: 4 weeks (8 sessions / 20 hours)

Goal: Build analytical thinking and statistics foundation before tools. Adding light visualization early so students see value in data storytelling immediately.

Week 1: Foundations of Analytics & Data Thinking

Getting familiarized with each other
Course logistics
What is data analytics? Real business use cases
Types of analytics (descriptive, predictive, prescriptive)
Data lifecycle and roles (Analyst, Engineer, Scientist)
Understanding data sources and structures
Hands-on: Map the data analytics process for a telecom or retail case

Week 2: Descriptive Statistics + Intro to Data Visualization

Types of data
Mean, median, mode
Variance, standard deviation
Histogram, bar chart, box plot
When to use each chart type
Hands-on: basic charts in Excel or Sheets
Hands-on: summarize dataset (pandas describe)

Week 3: Probability & Distributions + Visualizing Distributions

Probability basics
Normal, Binomial, Poisson distributions
CLT, z-scores, outliers
Visualizing distributions: hist, KDE, scatter, density plots
Hands-on: dice/coin simulations in Python or Excel

Week 4: Inference & Analytical Thinking + Visual Storytelling

Sampling methods
Confidence intervals, margin of error
Hypothesis testing (t-test, chi-square)
Interpreting p-values and effect size
Bias
Storytelling with charts
Mini project: GameCo analysis

SQL

Duration: 6 weeks (12 sessions / 30 hours)

Goal: Build solid SQL and database skills by designing schemas, querying data, performing analytical transformations, and working with real PostgreSQL environments to support end-to-end analytics workflows

Week 1: Environment Setup — Part 1

Install Docker Desktop
Pull & run PostgreSQL container
Install pgAdmin or configure psql
Install VS Code + DB/Git extensions
Install Git, create GitHub account
Create project folder structure
Initialize Git repo
Verify DB connection

Week 2: Environment Setup — Part 2 + SQL Kickoff

Postgres lifecycle in Docker
Basic PostgreSQL config
Schemas, tables
Insert sample rows
SELECT statements
Load CSV to DB

Week 3: Filtering, Aggregation, and Grouping

SELECT, WHERE, ORDER BY, DISTINCT
SUM, AVG, COUNT, MIN, MAX
GROUP BY, HAVING
CASE, COALESCE, CAST
Commit SQL to GitHub

Week 4: Joins, Subqueries, and CTEs

Keys & relationships
INNER, LEFT, RIGHT, FULL
Subqueries
CTEs for pipelines
Export results to CSV

Week 5: Window Functions & Analytical Scenarios

OVER() clause
RANK, ROW_NUMBER, DENSE_RANK
Running totals, moving averages, lag/lead
Telecom churn & retention cases
ERD + data dictionary

Week 6: Functions, Procedures, and Advanced Aggregations

Stored procedures & UDFs
Parameterized queries
CUBE, ROLLUP, GROUPING SETS
Materialized views

Python for Data Analytics

Duration: 7 weeks (14 sessions / 35 hours)

Goal: Learn to connect data sources, build analytical visualizations, and design interactive dashboards that communicate insights effectively for business decisions.

Week 1: Python Fundamentals & Environment Setup

Install Miniconda
Create environment
Install pandas, numpy, matplotlib, seaborn, psycopg2
Install Jupyter Notebook
Connect Python ↔︎ PostgreSQL
Python basics: syntax, loops, functions, types
Run notebooks and scripts

Week 2: pandas & Data Visualization

Load, clean, transform datasets
describe, info, shape, value_counts
Aggregation, filtering, merging
Visualizations: hist, bar, box, line charts

Week 3: A/B Testing & Regression

Experiments, control/treatment
Hypothesis testing, p-values
t-test in Python
Linear regression (scikit-learn)
R², coefficients, interpretation

Week 4: Clustering & Segmentation

K-Means clustering
Elbow method
Visualizing clusters
Business segmentation interpretation

Week 5: SQLAlchemy Integration

Connect Python & PostgreSQL via SQLAlchemy
Read SQL tables into pandas
Execute SQL from Python
Write processed data back to DB

Week 6: Streamlit Fundamentals + Prototyping

Intro to Streamlit: layout, widgets, state
Project structure for apps (app.py, pages/, assets/)
Connect to PostgreSQL from Streamlit (SQLAlchemy)
Display DataFrames, filters, and basic charts
Caching data queries and expensive operations
Prototype: single‑page KPI dashboard

Week 7: Final Streamlit Data Dashboard

Design multi‑page app (navigation via pages/)
Interactive filtering, parameters, and URL/query state
Charts with seaborn/matplotlib or Plotly; export to CSV
Session state, forms, and callbacks for smooth UX
Environment configs (.env) and secrets management
Optional deployment: Streamlit Community Cloud or Docker
Deliverable: Final dashboard demo with README instructions

Tableau

Duration: 4 weeks (8 sessions / 20 hours)

Goal: Develop hands-on expertise in data visualization and dashboard design, connecting to real data sources and building interactive, business-ready dashboards.

Week 1: Introduction to Tableau & Visual Analytics

Overview of Tableau and its role in data analysis and business intelligence
Connecting to data sources (Excel, CSV)
Understanding Tableau products and file types
Navigating the Tableau interface (Data Pane, Shelves, Marks Card, Filters, Dashboards)
Tableau data model: Data types, Dimensions vs. Measures, Discrete vs. Continuous fields
Creating basic visualizations: Bar charts, Line charts, Pie charts, Scatter plots
Adding interactivity with filters, groups, sets, and sorting
Hands-on task: Build and publish a simple Tableau dashboard using the Airbnb dataset

Week 2: Intermediate Visual Analytics

Connect to databases
Work with data joins, blends, unions, and relationships
Get CSV from SQL Stored Procedure using Python
Dual-axis charts, histograms, boxplots, heatmaps
Calculated fields (row-level vs aggregate)
Level of Detail (LOD) expressions (FIXED, INCLUDE, EXCLUDE)
Parameters (interactive filtering)
Types of filters (context, dimension, measure)
Task: Build dashboards with interactivity (actions, filters, tooltips)

Week 3: Advanced Analytics

Complex Calculations
Table calculations (running totals, percent of total, rank)
Date functions
Date parameters
Cohort and retention analysis in Tableau
Spatial Analytics (spatial relationships, spatial joins, spatial functions)
Mapping (map layers, map styling)
Data Modeling
Data prep with Tableau Prep
Cleaning and reshaping data
Build dashboards for KPIs, cohort tables, and advanced heatmaps
Work on telecom/finance/marketing datasets

Week 4: Dashboard Design & Performance

Design Principles
Visual best practices (color, layout, storytelling)
Types of dashboards
Interactive dashboards
Performance Optimization
Reduce extract size
Optimize calculations
Minimize dashboard load time
Task: Build an end-to-end business dashboard (filters, parameters, highlights) and add interactivity (URL actions, sheet swapping)

Capstone

Duration: 3 weeks (8 sessions / 15 hours)

Goal: Build and present a complete data analytics project demonstrating end-to-end skills across data modeling, ETL, SQL analysis, Python insights, and dashboard storytelling.

Problem Definition: Identify business question
Data Planning & Schema Design: Define ERD & tables
ETL using Python: Extract, clean, and load data to SQL
Analytical Layer (SQL): Create stored procedures, summary tables
Exploration & Predictive Analytics (Python): Analyze and model data
Export Final Data (CSV): prepare dataset for visualization
Dashboard Creation (Tableau): Final visual storytelling