World Robot Olympiad Competition Preparation Coaching in Pasir Ris

1. Making Sense of Data with Descriptive Statistics

Descriptive statistics summarize structure and check that values make sense before modeling. Useful tools include:

• df.describe() — multi-row summary across numeric features
• value_counts() — frequency of distinct values
• quantile() — split distributions into probability portions
• skew() — whether the curve leans left or right

Core ideas: central tendency (mean, median), dispersion (std, quartiles, IQR), shape (skewness), and simple validity checks (e.g. Q3 > Q1).

2. Quantiles and the Interquartile Range (IQR)

Quartiles each cover about 25% of observations:

• Q1 (25%) — lower quarter boundary
• Q2 (50% / median) — middle of the data
• Q3 (75%) — upper quarter boundary

IQR = Q3 − Q1 measures spread of the middle 50% of values.

3. Skewness

• Right-skewed (positive) — long right tail; mean > median; .skew() > 0
• Left-skewed (negative) — long left tail; mean < median; .skew() < 0

4. Practical Example: E-Commerce Revenue

import pandas as pd

df = pd.read_csv("ecommerce_orders.csv", parse_dates=["Order Date"])
print("Dataset Shape:", df.shape)

stats = df[["Unit Price", "Revenue"]].describe()
print(stats)

q1 = df["Revenue"].quantile(0.25)
q2 = df["Revenue"].quantile(0.50)
q3 = df["Revenue"].quantile(0.75)
print(f"Q1: {q1}, Median: {q2}, Q3: {q3}")

print("Mean:", df["Revenue"].mean())
print("Median:", df["Revenue"].median())
print("Skew:", df["Revenue"].skew())

print(df.groupby("Product Category")["Revenue"].mean())

iqr = q3 - q1
print("IQR:", iqr)
print("Q3 > Q1:", q3 > q1)
print("IQR > 0:", iqr > 0)

Illustrative results on a 5,000-row sample: Q1 ≈ 354, median ≈ 796, Q3 ≈ 1,515; mean ≈ 1,021 (above median → right skew ≈ 1.017); IQR ≈ 1,161.

5. Group Means and Validation

groupby(...).mean() compares average revenue by product category. Checking Q3 > Q1 and IQR > 0 confirms the quartile order is sensible.